1 ACPI ACPI support is enabled. 2 AGP AGP (Accelerated Graphics Port) is enabled. 3 ALSA ALSA sound support is enabled. 4 APIC APIC support is enabled. 5 APM Advanced Power Management support is enabled. 6 APPARMOR AppArmor support is enabled. 7 ARM ARM architecture is enabled. 8 ARM64 ARM64 architecture is enabled. 9 AX25 Appropriate AX.25 support is enabled. 10 CLK Common clock infrastructure is enabled. 11 CMA Contiguous Memory Area support is enabled. 12 DRM Direct Rendering Management support is enabled. 13 DYNAMIC_DEBUG Build in debug messages and enable them at runtime 14 EARLY Parameter processed too early to be embedded in initrd. 15 EDD BIOS Enhanced Disk Drive Services (EDD) is enabled 16 EFI EFI Partitioning (GPT) is enabled 17 EVM Extended Verification Module 18 FB The frame buffer device is enabled. 19 FTRACE Function tracing enabled. 20 GCOV GCOV profiling is enabled. 21 HIBERNATION HIBERNATION is enabled. 22 HW Appropriate hardware is enabled. 23 HYPER_V HYPERV support is enabled. 24 IMA Integrity measurement architecture is enabled. 25 IP_PNP IP DHCP, BOOTP, or RARP is enabled. 26 IPV6 IPv6 support is enabled. 27 ISAPNP ISA PnP code is enabled. 28 ISDN Appropriate ISDN support is enabled. 29 ISOL CPU Isolation is enabled. 30 JOY Appropriate joystick support is enabled. 31 KGDB Kernel debugger support is enabled. 32 KVM Kernel Virtual Machine support is enabled. 33 LIBATA Libata driver is enabled 34 LOONGARCH LoongArch architecture is enabled. 35 LOOP Loopback device support is enabled. 36 LP Printer support is enabled. 37 M68k M68k architecture is enabled. 38 These options have more detailed description inside of 39 Documentation/arch/m68k/kernel-options.rst. 40 MDA MDA console support is enabled. 41 MIPS MIPS architecture is enabled. 42 MOUSE Appropriate mouse support is enabled. 43 MSI Message Signaled Interrupts (PCI). 44 MTD MTD (Memory Technology Device) support is enabled. 45 NET Appropriate network support is enabled. 46 NFS Appropriate NFS support is enabled. 47 NUMA NUMA support is enabled. 48 OF Devicetree is enabled. 49 PARISC The PA-RISC architecture is enabled. 50 PCI PCI bus support is enabled. 51 PCIE PCI Express support is enabled. 52 PCMCIA The PCMCIA subsystem is enabled. 53 PNP Plug & Play support is enabled. 54 PPC PowerPC architecture is enabled. 55 PPT Parallel port support is enabled. 56 PS2 Appropriate PS/2 support is enabled. 57 PV_OPS A paravirtualized kernel is enabled. 58 RAM RAM disk support is enabled. 59 RDT Intel Resource Director Technology. 60 RISCV RISCV architecture is enabled. 61 S390 S390 architecture is enabled. 62 SCSI Appropriate SCSI support is enabled. 63 A lot of drivers have their options described inside 64 the Documentation/scsi/ sub-directory. 65 SDW SoundWire support is enabled. 66 SECURITY Different security models are enabled. 67 SELINUX SELinux support is enabled. 68 SERIAL Serial support is enabled. 69 SH SuperH architecture is enabled. 70 SMP The kernel is an SMP kernel. 71 SPARC Sparc architecture is enabled. 72 SUSPEND System suspend states are enabled. 73 SWSUSP Software suspend (hibernation) is enabled. 74 TPM TPM drivers are enabled. 75 UMS USB Mass Storage support is enabled. 76 USB USB support is enabled. 77 USBHID USB Human Interface Device support is enabled. 78 V4L Video For Linux support is enabled. 79 VGA The VGA console has been enabled. 80 VMMIO Driver for memory mapped virtio devices is enabled. 81 VT Virtual terminal support is enabled. 82 WDT Watchdog support is enabled. 83 X86-32 X86-32, aka i386 architecture is enabled. 84 X86-64 X86-64 architecture is enabled. 85 X86 Either 32-bit or 64-bit x86 (same as X86-32+X86-64) 86 X86_UV SGI UV support is enabled. 87 XEN Xen support is enabled 88 XTENSA xtensa architecture is enabled. 89 90In addition, the following text indicates that the option 91 92 BOOT Is a boot loader parameter. 93 BUGS= Relates to possible processor bugs on the said processor. 94 KNL Is a kernel start-up parameter. 95 96 97Kernel parameters 98 99 accept_memory= [MM] 100 Format: { eager | lazy } 101 default: lazy 102 By default, unaccepted memory is accepted lazily to 103 avoid prolonged boot times. The lazy option will add 104 some runtime overhead until all memory is eventually 105 accepted. In most cases the overhead is negligible. 106 For some workloads or for debugging purposes 107 accept_memory=eager can be used to accept all memory 108 at once during boot. 109 110 acpi= [HW,ACPI,X86,ARM64,RISCV64,EARLY] 111 Advanced Configuration and Power Interface 112 Format: { force | on | off | strict | noirq | rsdt | 113 copy_dsdt | nospcr } 114 force -- enable ACPI if default was off 115 on -- enable ACPI but allow fallback to DT [arm64,riscv64] 116 off -- disable ACPI if default was on 117 noirq -- do not use ACPI for IRQ routing 118 strict -- Be less tolerant of platforms that are not 119 strictly ACPI specification compliant. 120 rsdt -- prefer RSDT over (default) XSDT 121 copy_dsdt -- copy DSDT to memory 122 nocmcff -- Disable firmware first mode for corrected 123 errors. This disables parsing the HEST CMC error 124 source to check if firmware has set the FF flag. This 125 may result in duplicate corrected error reports. 126 nospcr -- disable console in ACPI SPCR table as 127 default _serial_ console on ARM64 128 For ARM64, ONLY "acpi=off", "acpi=on", "acpi=force" or 129 "acpi=nospcr" are available 130 For RISCV64, ONLY "acpi=off", "acpi=on" or "acpi=force" 131 are available 132 133 See also Documentation/power/runtime_pm.rst, pci=noacpi 134 135 acpi_apic_instance= [ACPI,IOAPIC,EARLY] 136 Format: <int> 137 2: use 2nd APIC table, if available 138 1,0: use 1st APIC table 139 default: 0 140 141 acpi_backlight= [HW,ACPI] 142 { vendor | video | native | none } 143 If set to vendor, prefer vendor-specific driver 144 (e.g. thinkpad_acpi, sony_acpi, etc.) instead 145 of the ACPI video.ko driver. 146 If set to video, use the ACPI video.ko driver. 147 If set to native, use the device's native backlight mode. 148 If set to none, disable the ACPI backlight interface. 149 150 acpi_force_32bit_fadt_addr [ACPI,EARLY] 151 force FADT to use 32 bit addresses rather than the 152 64 bit X_* addresses. Some firmware have broken 64 153 bit addresses for force ACPI ignore these and use 154 the older legacy 32 bit addresses. 155 156 acpica_no_return_repair [HW, ACPI] 157 Disable AML predefined validation mechanism 158 This mechanism can repair the evaluation result to make 159 the return objects more ACPI specification compliant. 160 This option is useful for developers to identify the 161 root cause of an AML interpreter issue when the issue 162 has something to do with the repair mechanism. 163 164 acpi.debug_layer= [HW,ACPI,ACPI_DEBUG] 165 acpi.debug_level= [HW,ACPI,ACPI_DEBUG] 166 Format: <int> 167 CONFIG_ACPI_DEBUG must be enabled to produce any ACPI 168 debug output. Bits in debug_layer correspond to a 169 _COMPONENT in an ACPI source file, e.g., 170 #define _COMPONENT ACPI_EVENTS 171 Bits in debug_level correspond to a level in 172 ACPI_DEBUG_PRINT statements, e.g., 173 ACPI_DEBUG_PRINT((ACPI_DB_INFO, ... 174 The debug_level mask defaults to "info". See 175 Documentation/firmware-guide/acpi/debug.rst for more information about 176 debug layers and levels. 177 178 Enable processor driver info messages: 179 acpi.debug_layer=0x20000000 180 Enable AML "Debug" output, i.e., stores to the Debug 181 object while interpreting AML: 182 acpi.debug_layer=0xffffffff acpi.debug_level=0x2 183 Enable all messages related to ACPI hardware: 184 acpi.debug_layer=0x2 acpi.debug_level=0xffffffff 185 186 Some values produce so much output that the system is 187 unusable. The "log_buf_len" parameter may be useful 188 if you need to capture more output. 189 190 acpi_enforce_resources= [ACPI] 191 { strict | lax | no } 192 Check for resource conflicts between native drivers 193 and ACPI OperationRegions (SystemIO and SystemMemory 194 only). IO ports and memory declared in ACPI might be 195 used by the ACPI subsystem in arbitrary AML code and 196 can interfere with legacy drivers. 197 strict (default): access to resources claimed by ACPI 198 is denied; legacy drivers trying to access reserved 199 resources will fail to bind to device using them. 200 lax: access to resources claimed by ACPI is allowed; 201 legacy drivers trying to access reserved resources 202 will bind successfully but a warning message is logged. 203 no: ACPI OperationRegions are not marked as reserved, 204 no further checks are performed. 205 206 acpi_force_table_verification [HW,ACPI,EARLY] 207 Enable table checksum verification during early stage. 208 By default, this is disabled due to x86 early mapping 209 size limitation. 210 211 acpi_irq_balance [HW,ACPI] 212 ACPI will balance active IRQs 213 default in APIC mode 214 215 acpi_irq_nobalance [HW,ACPI] 216 ACPI will not move active IRQs (default) 217 default in PIC mode 218 219 acpi_irq_isa= [HW,ACPI] If irq_balance, mark listed IRQs used by ISA 220 Format: <irq>,<irq>... 221 222 acpi_irq_pci= [HW,ACPI] If irq_balance, clear listed IRQs for 223 use by PCI 224 Format: <irq>,<irq>... 225 226 acpi_mask_gpe= [HW,ACPI] 227 Due to the existence of _Lxx/_Exx, some GPEs triggered 228 by unsupported hardware/firmware features can result in 229 GPE floodings that cannot be automatically disabled by 230 the GPE dispatcher. 231 This facility can be used to prevent such uncontrolled 232 GPE floodings. 233 Format: <byte> or <bitmap-list> 234 235 acpi_no_auto_serialize [HW,ACPI] 236 Disable auto-serialization of AML methods 237 AML control methods that contain the opcodes to create 238 named objects will be marked as "Serialized" by the 239 auto-serialization feature. 240 This feature is enabled by default. 241 This option allows to turn off the feature. 242 243 acpi_no_memhotplug [ACPI] Disable memory hotplug. Useful for kdump 244 kernels. 245 246 acpi_no_static_ssdt [HW,ACPI,EARLY] 247 Disable installation of static SSDTs at early boot time 248 By default, SSDTs contained in the RSDT/XSDT will be 249 installed automatically and they will appear under 250 /sys/firmware/acpi/tables. 251 This option turns off this feature. 252 Note that specifying this option does not affect 253 dynamic table installation which will install SSDT 254 tables to /sys/firmware/acpi/tables/dynamic. 255 256 acpi_no_watchdog [HW,ACPI,WDT] 257 Ignore the ACPI-based watchdog interface (WDAT) and let 258 a native driver control the watchdog device instead. 259 260 acpi_rsdp= [ACPI,EFI,KEXEC,EARLY] 261 Pass the RSDP address to the kernel, mostly used 262 on machines running EFI runtime service to boot the 263 second kernel for kdump. 264 265 acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS 266 Format: To spoof as Windows 98: ="Microsoft Windows" 267 268 acpi_rev_override [ACPI] Override the _REV object to return 5 (instead 269 of 2 which is mandated by ACPI 6) as the supported ACPI 270 specification revision (when using this switch, it may 271 be necessary to carry out a cold reboot _twice_ in a 272 row to make it take effect on the platform firmware). 273 274 acpi_osi= [HW,ACPI] Modify list of supported OS interface strings 275 acpi_osi="string1" # add string1 276 acpi_osi="!string2" # remove string2 277 acpi_osi=!* # remove all strings 278 acpi_osi=! # disable all built-in OS vendor 279 strings 280 acpi_osi=!! # enable all built-in OS vendor 281 strings 282 acpi_osi= # disable all strings 283 284 'acpi_osi=!' can be used in combination with single or 285 multiple 'acpi_osi="string1"' to support specific OS 286 vendor string(s). Note that such command can only 287 affect the default state of the OS vendor strings, thus 288 it cannot affect the default state of the feature group 289 strings and the current state of the OS vendor strings, 290 specifying it multiple times through kernel command line 291 is meaningless. This command is useful when one do not 292 care about the state of the feature group strings which 293 should be controlled by the OSPM. 294 Examples: 295 1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent 296 to 'acpi_osi="Windows 2000" acpi_osi=!', they all 297 can make '_OSI("Windows 2000")' TRUE. 298 299 'acpi_osi=' cannot be used in combination with other 300 'acpi_osi=' command lines, the _OSI method will not 301 exist in the ACPI namespace. NOTE that such command can 302 only affect the _OSI support state, thus specifying it 303 multiple times through kernel command line is also 304 meaningless. 305 Examples: 306 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)' 307 FALSE. 308 309 'acpi_osi=!*' can be used in combination with single or 310 multiple 'acpi_osi="string1"' to support specific 311 string(s). Note that such command can affect the 312 current state of both the OS vendor strings and the 313 feature group strings, thus specifying it multiple times 314 through kernel command line is meaningful. But it may 315 still not able to affect the final state of a string if 316 there are quirks related to this string. This command 317 is useful when one want to control the state of the 318 feature group strings to debug BIOS issues related to 319 the OSPM features. 320 Examples: 321 1. 'acpi_osi="Module Device" acpi_osi=!*' can make 322 '_OSI("Module Device")' FALSE. 323 2. 'acpi_osi=!* acpi_osi="Module Device"' can make 324 '_OSI("Module Device")' TRUE. 325 3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is 326 equivalent to 327 'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"' 328 and 329 'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!', 330 they all will make '_OSI("Windows 2000")' TRUE. 331 332 acpi_pm_good [X86] 333 Override the pmtimer bug detection: force the kernel 334 to assume that this machine's pmtimer latches its value 335 and always returns good values. 336 337 acpi_sci= [HW,ACPI,EARLY] ACPI System Control Interrupt trigger mode 338 Format: { level | edge | high | low } 339 340 acpi_skip_timer_override [HW,ACPI,EARLY] 341 Recognize and ignore IRQ0/pin2 Interrupt Override. 342 For broken nForce2 BIOS resulting in XT-PIC timer. 343 344 acpi_sleep= [HW,ACPI] Sleep options 345 Format: { s3_bios, s3_mode, s3_beep, s4_hwsig, 346 s4_nohwsig, old_ordering, nonvs, 347 sci_force_enable, nobl } 348 See Documentation/power/video.rst for information on 349 s3_bios and s3_mode. 350 s3_beep is for debugging; it makes the PC's speaker beep 351 as soon as the kernel's real-mode entry point is called. 352 s4_hwsig causes the kernel to check the ACPI hardware 353 signature during resume from hibernation, and gracefully 354 refuse to resume if it has changed. This complies with 355 the ACPI specification but not with reality, since 356 Windows does not do this and many laptops do change it 357 on docking. So the default behaviour is to allow resume 358 and simply warn when the signature changes, unless the 359 s4_hwsig option is enabled. 360 s4_nohwsig prevents ACPI hardware signature from being 361 used (or even warned about) during resume. 362 old_ordering causes the ACPI 1.0 ordering of the _PTS 363 control method, with respect to putting devices into 364 low power states, to be enforced (the ACPI 2.0 ordering 365 of _PTS is used by default). 366 nonvs prevents the kernel from saving/restoring the 367 ACPI NVS memory during suspend/hibernation and resume. 368 sci_force_enable causes the kernel to set SCI_EN directly 369 on resume from S1/S3 (which is against the ACPI spec, 370 but some broken systems don't work without it). 371 nobl causes the internal blacklist of systems known to 372 behave incorrectly in some ways with respect to system 373 suspend and resume to be ignored (use wisely). 374 375 acpi_use_timer_override [HW,ACPI,EARLY] 376 Use timer override. For some broken Nvidia NF5 boards 377 that require a timer override, but don't have HPET 378 379 add_efi_memmap [EFI,X86,EARLY] Include EFI memory map in 380 kernel's map of available physical RAM. 381 382 agp= [AGP] 383 { off | try_unsupported } 384 off: disable AGP support 385 try_unsupported: try to drive unsupported chipsets 386 (may crash computer or cause data corruption) 387 388 ALSA [HW,ALSA] 389 See Documentation/sound/alsa-configuration.rst 390 391 alignment= [KNL,ARM] 392 Allow the default userspace alignment fault handler 393 behaviour to be specified. Bit 0 enables warnings, 394 bit 1 enables fixups, and bit 2 sends a segfault. 395 396 align_va_addr= [X86-64] 397 Align virtual addresses by clearing slice [14:12] when 398 allocating a VMA at process creation time. This option 399 gives you up to 3% performance improvement on AMD F15h 400 machines (where it is enabled by default) for a 401 CPU-intensive style benchmark, and it can vary highly in 402 a microbenchmark depending on workload and compiler. 403 404 32: only for 32-bit processes 405 64: only for 64-bit processes 406 on: enable for both 32- and 64-bit processes 407 off: disable for both 32- and 64-bit processes 408 409 alloc_snapshot [FTRACE] 410 Allocate the ftrace snapshot buffer on boot up when the 411 main buffer is allocated. This is handy if debugging 412 and you need to use tracing_snapshot() on boot up, and 413 do not want to use tracing_snapshot_alloc() as it needs 414 to be done where GFP_KERNEL allocations are allowed. 415 416 allow_mismatched_32bit_el0 [ARM64,EARLY] 417 Allow execve() of 32-bit applications and setting of the 418 PER_LINUX32 personality on systems where only a strict 419 subset of the CPUs support 32-bit EL0. When this 420 parameter is present, the set of CPUs supporting 32-bit 421 EL0 is indicated by /sys/devices/system/cpu/aarch32_el0 422 and hot-unplug operations may be restricted. 423 424 See Documentation/arch/arm64/asymmetric-32bit.rst for more 425 information. 426 427 amd_iommu= [HW,X86-64] 428 Pass parameters to the AMD IOMMU driver in the system. 429 Possible values are: 430 fullflush - Deprecated, equivalent to iommu.strict=1 431 off - do not initialize any AMD IOMMU found in 432 the system 433 force_isolation - Force device isolation for all 434 devices. The IOMMU driver is not 435 allowed anymore to lift isolation 436 requirements as needed. This option 437 does not override iommu=pt 438 force_enable - Force enable the IOMMU on platforms known 439 to be buggy with IOMMU enabled. Use this 440 option with care. 441 pgtbl_v1 - Use v1 page table for DMA-API (Default). 442 pgtbl_v2 - Use v2 page table for DMA-API. 443 irtcachedis - Disable Interrupt Remapping Table (IRT) caching. 444 nohugepages - Limit page-sizes used for v1 page-tables 445 to 4 KiB. 446 v2_pgsizes_only - Limit page-sizes used for v1 page-tables 447 to 4KiB/2Mib/1GiB. 448 449 450 amd_iommu_dump= [HW,X86-64] 451 Enable AMD IOMMU driver option to dump the ACPI table 452 for AMD IOMMU. With this option enabled, AMD IOMMU 453 driver will print ACPI tables for AMD IOMMU during 454 IOMMU initialization. 455 456 amd_iommu_intr= [HW,X86-64] 457 Specifies one of the following AMD IOMMU interrupt 458 remapping modes: 459 legacy - Use legacy interrupt remapping mode. 460 vapic - Use virtual APIC mode, which allows IOMMU 461 to inject interrupts directly into guest. 462 This mode requires kvm-amd.avic=1. 463 (Default when IOMMU HW support is present.) 464 465 amd_pstate= [X86,EARLY] 466 disable 467 Do not enable amd_pstate as the default 468 scaling driver for the supported processors 469 passive 470 Use amd_pstate with passive mode as a scaling driver. 471 In this mode autonomous selection is disabled. 472 Driver requests a desired performance level and platform 473 tries to match the same performance level if it is 474 satisfied by guaranteed performance level. 475 active 476 Use amd_pstate_epp driver instance as the scaling driver, 477 driver provides a hint to the hardware if software wants 478 to bias toward performance (0x0) or energy efficiency (0xff) 479 to the CPPC firmware. then CPPC power algorithm will 480 calculate the runtime workload and adjust the realtime cores 481 frequency. 482 guided 483 Activate guided autonomous mode. Driver requests minimum and 484 maximum performance level and the platform autonomously 485 selects a performance level in this range and appropriate 486 to the current workload. 487 488 amd_prefcore= 489 [X86] 490 disable 491 Disable amd-pstate preferred core. 492 493 amijoy.map= [HW,JOY] Amiga joystick support 494 Map of devices attached to JOY0DAT and JOY1DAT 495 Format: <a>,<b> 496 See also Documentation/input/joydev/joystick.rst 497 498 analog.map= [HW,JOY] Analog joystick and gamepad support 499 Specifies type or capabilities of an analog joystick 500 connected to one of 16 gameports 501 Format: <type1>,<type2>,..<type16> 502 503 apc= [HW,SPARC] 504 Power management functions (SPARCstation-4/5 + deriv.) 505 Format: noidle 506 Disable APC CPU standby support. SPARCstation-Fox does 507 not play well with APC CPU idle - disable it if you have 508 APC and your system crashes randomly. 509 510 apic [APIC,X86-64] Use IO-APIC. Default. 511 512 apic= [APIC,X86,EARLY] Advanced Programmable Interrupt Controller 513 Change the output verbosity while booting 514 Format: { quiet (default) | verbose | debug } 515 Change the amount of debugging information output 516 when initialising the APIC and IO-APIC components. 517 518 apic_extnmi= [APIC,X86,EARLY] External NMI delivery setting 519 Format: { bsp (default) | all | none } 520 bsp: External NMI is delivered only to CPU 0 521 all: External NMIs are broadcast to all CPUs as a 522 backup of CPU 0 523 none: External NMI is masked for all CPUs. This is 524 useful so that a dump capture kernel won't be 525 shot down by NMI 526 527 apicpmtimer Do APIC timer calibration using the pmtimer. Implies 528 apicmaintimer. Useful when your PIT timer is totally 529 broken. 530 531 autoconf= [IPV6] 532 See Documentation/networking/ipv6.rst. 533 534 apm= [APM] Advanced Power Management 535 See header of arch/x86/kernel/apm_32.c. 536 537 apparmor= [APPARMOR] Disable or enable AppArmor at boot time 538 Format: { "0" | "1" } 539 See security/apparmor/Kconfig help text 540 0 -- disable. 541 1 -- enable. 542 Default value is set via kernel config option. 543 544 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards 545 Format: <io>,<irq>,<nodeID> 546 547 arm64.no32bit_el0 [ARM64] Unconditionally disable the execution of 548 32 bit applications. 549 550 arm64.nobti [ARM64] Unconditionally disable Branch Target 551 Identification support 552 553 arm64.nogcs [ARM64] Unconditionally disable Guarded Control Stack 554 support 555 556 arm64.nomops [ARM64] Unconditionally disable Memory Copy and Memory 557 Set instructions support 558 559 arm64.nompam [ARM64] Unconditionally disable Memory Partitioning And 560 Monitoring support 561 562 arm64.nomte [ARM64] Unconditionally disable Memory Tagging Extension 563 support 564 565 arm64.nopauth [ARM64] Unconditionally disable Pointer Authentication 566 support 567 568 arm64.nosme [ARM64] Unconditionally disable Scalable Matrix 569 Extension support 570 571 arm64.nosve [ARM64] Unconditionally disable Scalable Vector 572 Extension support 573 574 ataflop= [HW,M68k] 575 576 atarimouse= [HW,MOUSE] Atari Mouse 577 578 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess, 579 EzKey and similar keyboards 580 581 atkbd.reset= [HW] Reset keyboard during initialization 582 583 atkbd.set= [HW] Select keyboard code set 584 Format: <int> (2 = AT (default), 3 = PS/2) 585 586 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar 587 keyboards 588 589 atkbd.softraw= [HW] Choose between synthetic and real raw mode 590 Format: <bool> (0 = real, 1 = synthetic (default)) 591 592 atkbd.softrepeat= [HW] 593 Use software keyboard repeat 594 595 audit= [KNL] Enable the audit sub-system 596 Format: { "0" | "1" | "off" | "on" } 597 0 | off - kernel audit is disabled and can not be 598 enabled until the next reboot 599 unset - kernel audit is initialized but disabled and 600 will be fully enabled by the userspace auditd. 601 1 | on - kernel audit is initialized and partially 602 enabled, storing at most audit_backlog_limit 603 messages in RAM until it is fully enabled by the 604 userspace auditd. 605 Default: unset 606 607 audit_backlog_limit= [KNL] Set the audit queue size limit. 608 Format: <int> (must be >=0) 609 Default: 64 610 611 bau= [X86_UV] Enable the BAU on SGI UV. The default 612 behavior is to disable the BAU (i.e. bau=0). 613 Format: { "0" | "1" } 614 0 - Disable the BAU. 615 1 - Enable the BAU. 616 unset - Disable the BAU. 617 618 baycom_epp= [HW,AX25] 619 Format: <io>,<mode> 620 621 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem 622 Format: <io>,<mode> 623 See header of drivers/net/hamradio/baycom_par.c. 624 625 baycom_ser_fdx= [HW,AX25] 626 BayCom Serial Port AX.25 Modem (Full Duplex Mode) 627 Format: <io>,<irq>,<mode>[,<baud>] 628 See header of drivers/net/hamradio/baycom_ser_fdx.c. 629 630 baycom_ser_hdx= [HW,AX25] 631 BayCom Serial Port AX.25 Modem (Half Duplex Mode) 632 Format: <io>,<irq>,<mode> 633 See header of drivers/net/hamradio/baycom_ser_hdx.c. 634 635 bdev_allow_write_mounted= 636 Format: <bool> 637 Control the ability to open a mounted block device 638 for writing, i.e., allow / disallow writes that bypass 639 the FS. This was implemented as a means to prevent 640 fuzzers from crashing the kernel by overwriting the 641 metadata underneath a mounted FS without its awareness. 642 This also prevents destructive formatting of mounted 643 filesystems by naive storage tooling that don't use 644 O_EXCL. Default is Y and can be changed through the 645 Kconfig option CONFIG_BLK_DEV_WRITE_MOUNTED. 646 647 bert_disable [ACPI] 648 Disable BERT OS support on buggy BIOSes. 649 650 bgrt_disable [ACPI,X86,EARLY] 651 Disable BGRT to avoid flickering OEM logo. 652 653 blkdevparts= Manual partition parsing of block device(s) for 654 embedded devices based on command line input. 655 See Documentation/block/cmdline-partition.rst 656 657 boot_delay= [KNL,EARLY] 658 Milliseconds to delay each printk during boot. 659 Only works if CONFIG_BOOT_PRINTK_DELAY is enabled, 660 and you may also have to specify "lpj=". Boot_delay 661 values larger than 10 seconds (10000) are assumed 662 erroneous and ignored. 663 Format: integer 664 665 bootconfig [KNL,EARLY] 666 Extended command line options can be added to an initrd 667 and this will cause the kernel to look for it. 668 669 See Documentation/admin-guide/bootconfig.rst 670 671 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards) 672 bttv.radio= Most important insmod options are available as 673 kernel args too. 674 bttv.pll= See Documentation/admin-guide/media/bttv.rst 675 bttv.tuner= 676 677 bulk_remove=off [PPC] This parameter disables the use of the pSeries 678 firmware feature for flushing multiple hpte entries 679 at a time. 680 681 c101= [NET] Moxa C101 synchronous serial card 682 683 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection. 684 Sometimes CPU hardware bugs make them report the cache 685 size incorrectly. The kernel will attempt work arounds 686 to fix known problems, but for some CPUs it is not 687 possible to determine what the correct size should be. 688 This option provides an override for these situations. 689 690 carrier_timeout= 691 [NET] Specifies amount of time (in seconds) that 692 the kernel should wait for a network carrier. By default 693 it waits 120 seconds. 694 695 ca_keys= [KEYS] This parameter identifies a specific key(s) on 696 the system trusted keyring to be used for certificate 697 trust validation. 698 format: { id:<keyid> | builtin } 699 700 cca= [MIPS,EARLY] Override the kernel pages' cache coherency 701 algorithm. Accepted values range from 0 to 7 702 inclusive. See arch/mips/include/asm/pgtable-bits.h 703 for platform specific values (SB1, Loongson3 and 704 others). 705 706 ccw_timeout_log [S390] 707 See Documentation/arch/s390/common_io.rst for details. 708 709 cfi= [X86-64] Set Control Flow Integrity checking features 710 when CONFIG_FINEIBT is enabled. 711 Format: feature[,feature...] 712 Default: auto 713 714 auto: Use FineIBT if IBT available, otherwise kCFI. 715 Under FineIBT, enable "paranoid" mode when 716 FRED is not available. 717 off: Turn off CFI checking. 718 kcfi: Use kCFI (disable FineIBT). 719 fineibt: Use FineIBT (even if IBT not available). 720 norand: Do not re-randomize CFI hashes. 721 paranoid: Add caller hash checking under FineIBT. 722 bhi: Enable register poisoning to stop speculation 723 across FineIBT. (Disabled by default.) 724 warn: Do not enforce CFI checking: warn only. 725 debug: Report CFI initialization details. 726 727 cgroup_disable= [KNL] Disable a particular controller or optional feature 728 Format: {name of the controller(s) or feature(s) to disable} 729 The effects of cgroup_disable=foo are: 730 - foo isn't auto-mounted if you mount all cgroups in 731 a single hierarchy 732 - foo isn't visible as an individually mountable 733 subsystem 734 - if foo is an optional feature then the feature is 735 disabled and corresponding cgroup files are not 736 created 737 {Currently only "memory" controller deal with this and 738 cut the overhead, others just disable the usage. So 739 only cgroup_disable=memory is actually worthy} 740 Specifying "pressure" disables per-cgroup pressure 741 stall information accounting feature 742 743 cgroup_no_v1= [KNL] Disable cgroup controllers and named hierarchies in v1 744 Format: { { controller | "all" | "named" } 745 [,{ controller | "all" | "named" }...] } 746 Like cgroup_disable, but only applies to cgroup v1; 747 the blacklisted controllers remain available in cgroup2. 748 "all" blacklists all controllers and "named" disables 749 named mounts. Specifying both "all" and "named" disables 750 all v1 hierarchies. 751 752 cgroup_v1_proc= [KNL] Show also missing controllers in /proc/cgroups 753 Format: { "true" | "false" } 754 /proc/cgroups lists only v1 controllers by default. 755 This compatibility option enables listing also v2 756 controllers (whose v1 code is not compiled!), so that 757 semi-legacy software can check this file to decide 758 about usage of v2 (sic) controllers. 759 760 cgroup_favordynmods= [KNL] Enable or Disable favordynmods. 761 Format: { "true" | "false" } 762 Defaults to the value of CONFIG_CGROUP_FAVOR_DYNMODS. 763 764 cgroup.memory= [KNL] Pass options to the cgroup memory controller. 765 Format: <string> 766 nosocket -- Disable socket memory accounting. 767 nokmem -- Disable kernel memory accounting. 768 nobpf -- Disable BPF memory accounting. 769 770 check_pages= [MM,EARLY] Enable sanity checking of pages after 771 allocations / before freeing. This adds checks to catch 772 double-frees, use-after-frees, and other sources of 773 page corruption by inspecting page internals (flags, 774 mapcount/refcount, memcg_data, etc.). 775 Format: { "0" | "1" } 776 Default: 0 (1 if CONFIG_DEBUG_VM is set) 777 778 checkreqprot= [SELINUX] Set initial checkreqprot flag value. 779 Format: { "0" | "1" } 780 See security/selinux/Kconfig help text. 781 0 -- check protection applied by kernel (includes 782 any implied execute protection). 783 1 -- check protection requested by application. 784 Default value is set via a kernel config option. 785 Value can be changed at runtime via 786 /sys/fs/selinux/checkreqprot. 787 Setting checkreqprot to 1 is deprecated. 788 789 cio_ignore= [S390] 790 See Documentation/arch/s390/common_io.rst for details. 791 792 clearcpuid=X[,X...] [X86] 793 Disable CPUID feature X for the kernel. See 794 arch/x86/include/asm/cpufeatures.h for the valid bit 795 numbers X. Note the Linux-specific bits are not necessarily 796 stable over kernel options, but the vendor-specific 797 ones should be. 798 X can also be a string as appearing in the flags: line 799 in /proc/cpuinfo which does not have the above 800 instability issue. However, not all features have names 801 in /proc/cpuinfo. 802 Note that using this option will taint your kernel. 803 Also note that user programs calling CPUID directly 804 or using the feature without checking anything 805 will still see it. This just prevents it from 806 being used by the kernel or shown in /proc/cpuinfo. 807 Also note the kernel might malfunction if you disable 808 some critical bits. 809 810 clk_ignore_unused 811 [CLK] 812 Prevents the clock framework from automatically gating 813 clocks that have not been explicitly enabled by a Linux 814 device driver but are enabled in hardware at reset or 815 by the bootloader/firmware. Note that this does not 816 force such clocks to be always-on nor does it reserve 817 those clocks in any way. This parameter is useful for 818 debug and development, but should not be needed on a 819 platform with proper driver support. For more 820 information, see Documentation/driver-api/clk.rst. 821 822 clock= [BUGS=X86-32, HW] gettimeofday clocksource override. 823 [Deprecated] 824 Forces specified clocksource (if available) to be used 825 when calculating gettimeofday(). If specified 826 clocksource is not available, it defaults to PIT. 827 Format: { pit | tsc | cyclone | pmtmr } 828 829 clocksource= Override the default clocksource 830 Format: <string> 831 Override the default clocksource and use the clocksource 832 with the name specified. 833 Some clocksource names to choose from, depending on 834 the platform: 835 [all] jiffies (this is the base, fallback clocksource) 836 [ACPI] acpi_pm 837 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2, 838 pxa_timer,timer3,32k_counter,timer0_1 839 [X86-32] pit,hpet,tsc; 840 scx200_hrt on Geode; cyclone on IBM x440 841 [MIPS] MIPS 842 [PARISC] cr16 843 [S390] tod 844 [SH] SuperH 845 [SPARC64] tick 846 [X86-64] hpet,tsc 847 848 clocksource.arm_arch_timer.evtstrm= 849 [ARM,ARM64,EARLY] 850 Format: <bool> 851 Enable/disable the eventstream feature of the ARM 852 architected timer so that code using WFE-based polling 853 loops can be debugged more effectively on production 854 systems. 855 856 clocksource.verify_n_cpus= [KNL] 857 Limit the number of CPUs checked for clocksources 858 marked with CLOCK_SOURCE_VERIFY_PERCPU that 859 are marked unstable due to excessive skew. 860 A negative value says to check all CPUs, while 861 zero says not to check any. Values larger than 862 nr_cpu_ids are silently truncated to nr_cpu_ids. 863 The actual CPUs are chosen randomly, with 864 no replacement if the same CPU is chosen twice. 865 866 clocksource-wdtest.holdoff= [KNL] 867 Set the time in seconds that the clocksource 868 watchdog test waits before commencing its tests. 869 Defaults to zero when built as a module and to 870 10 seconds when built into the kernel. 871 872 cma=nn[MG]@[start[MG][-end[MG]]] 873 [KNL,CMA,EARLY] 874 Sets the size of kernel global memory area for 875 contiguous memory allocations and optionally the 876 placement constraint by the physical address range of 877 memory allocations. A value of 0 disables CMA 878 altogether. For more information, see 879 kernel/dma/contiguous.c 880 881 cma_pernuma=nn[MG] 882 [KNL,CMA,EARLY] 883 Sets the size of kernel per-numa memory area for 884 contiguous memory allocations. A value of 0 disables 885 per-numa CMA altogether. And If this option is not 886 specified, the default value is 0. 887 With per-numa CMA enabled, DMA users on node nid will 888 first try to allocate buffer from the pernuma area 889 which is located in node nid, if the allocation fails, 890 they will fallback to the global default memory area. 891 892 numa_cma=<node>:nn[MG][,<node>:nn[MG]] 893 [KNL,CMA,EARLY] 894 Sets the size of kernel numa memory area for 895 contiguous memory allocations. It will reserve CMA 896 area for the specified node. 897 898 With numa CMA enabled, DMA users on node nid will 899 first try to allocate buffer from the numa area 900 which is located in node nid, if the allocation fails, 901 they will fallback to the global default memory area. 902 903 cmo_free_hint= [PPC] Format: { yes | no } 904 Specify whether pages are marked as being inactive 905 when they are freed. This is used in CMO environments 906 to determine OS memory pressure for page stealing by 907 a hypervisor. 908 Default: yes 909 910 coherent_pool=nn[KMG] [ARM,KNL,EARLY] 911 Sets the size of memory pool for coherent, atomic dma 912 allocations, by default set to 256K. 913 914 com20020= [HW,NET] ARCnet - COM20020 chipset 915 Format: 916 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]] 917 918 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers) 919 Format: <io>[,<irq>] 920 921 com90xx= [HW,NET] 922 ARCnet - COM90xx chipset (memory-mapped buffers) 923 Format: <io>[,<irq>[,<memstart>]] 924 925 condev= [HW,S390] console device 926 conmode= 927 928 con3215_drop= [S390,EARLY] 3215 console drop mode. 929 Format: y|n|Y|N|1|0 930 When set to true, drop data on the 3215 console when 931 the console buffer is full. In this case the 932 operator using a 3270 terminal emulator (for example 933 x3270) does not have to enter the clear key for the 934 console output to advance and the kernel to continue. 935 This leads to a much faster boot time when a 3270 936 terminal emulator is active. If no 3270 terminal 937 emulator is used, this parameter has no effect. 938 939 console= [KNL] Output console device and options. 940 941 tty<n> Use the virtual console device <n>. 942 943 ttyS<n>[,options] 944 ttyUSB0[,options] 945 Use the specified serial port. The options are of 946 the form "bbbbpnf", where "bbbb" is the baud rate, 947 "p" is parity ("n", "o", or "e"), "n" is number of 948 bits, and "f" is flow control ("r" for RTS or 949 omit it). Default is "9600n8". 950 951 See Documentation/admin-guide/serial-console.rst for more 952 information. See 953 Documentation/networking/netconsole.rst for an 954 alternative. 955 956 <DEVNAME>:<n>.<n>[,options] 957 Use the specified serial port on the serial core bus. 958 The addressing uses DEVNAME of the physical serial port 959 device, followed by the serial core controller instance, 960 and the serial port instance. The options are the same 961 as documented for the ttyS addressing above. 962 963 The mapping of the serial ports to the tty instances 964 can be viewed with: 965 966 $ ls -d /sys/bus/serial-base/devices/*:*.*/tty/* 967 /sys/bus/serial-base/devices/00:04:0.0/tty/ttyS0 968 969 In the above example, the console can be addressed with 970 console=00:04:0.0. Note that a console addressed this 971 way will only get added when the related device driver 972 is ready. The use of an earlycon parameter in addition to 973 the console may be desired for console output early on. 974 975 uart[8250],io,<addr>[,options] 976 uart[8250],mmio,<addr>[,options] 977 uart[8250],mmio16,<addr>[,options] 978 uart[8250],mmio32,<addr>[,options] 979 uart[8250],0x<addr>[,options] 980 Start an early, polled-mode console on the 8250/16550 981 UART at the specified I/O port or MMIO address, 982 switching to the matching ttyS device later. 983 MMIO inter-register address stride is either 8-bit 984 (mmio), 16-bit (mmio16), or 32-bit (mmio32). 985 If none of [io|mmio|mmio16|mmio32], <addr> is assumed 986 to be equivalent to 'mmio'. 'options' are specified in 987 the same format described for ttyS above; if unspecified, 988 the h/w is not re-initialized. 989 990 hvc<n> Use the hypervisor console device <n>. This is for 991 both Xen and PowerPC hypervisors. 992 993 { null | "" } 994 Use to disable console output, i.e., to have kernel 995 console messages discarded. 996 This must be the only console= parameter used on the 997 kernel command line. 998 999 If the device connected to the port is not a TTY but a braille 1000 device, prepend "brl," before the device type, for instance 1001 console=brl,ttyS0 1002 For now, only VisioBraille is supported. 1003 1004 console_msg_format= 1005 [KNL] Change console messages format 1006 default 1007 By default we print messages on consoles in 1008 "[time stamp] text\n" format (time stamp may not be 1009 printed, depending on CONFIG_PRINTK_TIME or 1010 `printk_time' param). 1011 syslog 1012 Switch to syslog format: "<%u>[time stamp] text\n" 1013 IOW, each message will have a facility and loglevel 1014 prefix. The format is similar to one used by syslog() 1015 syscall, or to executing "dmesg -S --raw" or to reading 1016 from /proc/kmsg. 1017 1018 consoleblank= [KNL] The console blank (screen saver) timeout in 1019 seconds. A value of 0 disables the blank timer. 1020 Defaults to 0. 1021 1022 coredump_filter= 1023 [KNL] Change the default value for 1024 /proc/<pid>/coredump_filter. 1025 See also Documentation/filesystems/proc.rst. 1026 1027 coresight_cpu_debug.enable 1028 [ARM,ARM64] 1029 Format: <bool> 1030 Enable/disable the CPU sampling based debugging. 1031 0: default value, disable debugging 1032 1: enable debugging at boot time 1033 1034 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver 1035 Format: 1036 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>] 1037 1038 cpuidle.off=1 [CPU_IDLE] 1039 disable the cpuidle sub-system 1040 1041 cpuidle.governor= 1042 [CPU_IDLE] Name of the cpuidle governor to use. 1043 1044 cpufreq.off=1 [CPU_FREQ] 1045 disable the cpufreq sub-system 1046 1047 cpufreq.default_governor= 1048 [CPU_FREQ] Name of the default cpufreq governor or 1049 policy to use. This governor must be registered in the 1050 kernel before the cpufreq driver probes. 1051 1052 cpu_init_udelay=N 1053 [X86,EARLY] Delay for N microsec between assert and de-assert 1054 of APIC INIT to start processors. This delay occurs 1055 on every CPU online, such as boot, and resume from suspend. 1056 Default: 10000 1057 1058 cpuhp.parallel= 1059 [SMP] Enable/disable parallel bringup of secondary CPUs 1060 Format: <bool> 1061 Default is enabled if CONFIG_HOTPLUG_PARALLEL=y. Otherwise 1062 the parameter has no effect. 1063 1064 crash_kexec_post_notifiers 1065 Only jump to kdump kernel after running the panic 1066 notifiers and dumping kmsg. This option increases 1067 the risks of a kdump failure, since some panic 1068 notifiers can make the crashed kernel more unstable. 1069 In configurations where kdump may not be reliable, 1070 running the panic notifiers could allow collecting 1071 more data on dmesg, like stack traces from other CPUS 1072 or extra data dumped by panic_print. Note that some 1073 configurations enable this option unconditionally, 1074 like Hyper-V, PowerPC (fadump) and AMD SEV-SNP. 1075 1076 crashkernel=size[KMG][@offset[KMG]] 1077 [KNL,EARLY] Using kexec, Linux can switch to a 'crash kernel' 1078 upon panic. This parameter reserves the physical 1079 memory region [offset, offset + size] for that kernel 1080 image. If '@offset' is omitted, then a suitable offset 1081 is selected automatically. 1082 [KNL, X86-64, ARM64, RISCV, LoongArch] Select a region 1083 under 4G first, and fall back to reserve region above 1084 4G when '@offset' hasn't been specified. 1085 See Documentation/admin-guide/kdump/kdump.rst for further details. 1086 1087 crashkernel=range1:size1[,range2:size2,...][@offset] 1088 [KNL] Same as above, but depends on the memory 1089 in the running system. The syntax of range is 1090 start-[end] where start and end are both 1091 a memory unit (amount[KMG]). See also 1092 Documentation/admin-guide/kdump/kdump.rst for an example. 1093 1094 crashkernel=size[KMG],high 1095 [KNL, X86-64, ARM64, RISCV, LoongArch] range could be 1096 above 4G. 1097 Allow kernel to allocate physical memory region from top, 1098 so could be above 4G if system have more than 4G ram 1099 installed. Otherwise memory region will be allocated 1100 below 4G, if available. 1101 It will be ignored if crashkernel=X is specified. 1102 crashkernel=size[KMG],low 1103 [KNL, X86-64, ARM64, RISCV, LoongArch] range under 4G. 1104 When crashkernel=X,high is passed, kernel could allocate 1105 physical memory region above 4G, that cause second kernel 1106 crash on system that require some amount of low memory, 1107 e.g. swiotlb requires at least 64M+32K low memory, also 1108 enough extra low memory is needed to make sure DMA buffers 1109 for 32-bit devices won't run out. Kernel would try to allocate 1110 default size of memory below 4G automatically. The default 1111 size is platform dependent. 1112 --> x86: max(swiotlb_size_or_default() + 8MiB, 256MiB) 1113 --> arm64: 128MiB 1114 --> riscv: 128MiB 1115 --> loongarch: 128MiB 1116 This one lets the user specify own low range under 4G 1117 for second kernel instead. 1118 0: to disable low allocation. 1119 It will be ignored when crashkernel=X,high is not used 1120 or memory reserved is below 4G. 1121 crashkernel=size[KMG],cma 1122 [KNL, X86, ppc] Reserve additional crash kernel memory from 1123 CMA. This reservation is usable by the first system's 1124 userspace memory and kernel movable allocations (memory 1125 balloon, zswap). Pages allocated from this memory range 1126 will not be included in the vmcore so this should not 1127 be used if dumping of userspace memory is intended and 1128 it has to be expected that some movable kernel pages 1129 may be missing from the dump. 1130 1131 A standard crashkernel reservation, as described above, 1132 is still needed to hold the crash kernel and initrd. 1133 1134 This option increases the risk of a kdump failure: DMA 1135 transfers configured by the first kernel may end up 1136 corrupting the second kernel's memory. 1137 1138 This reservation method is intended for systems that 1139 can't afford to sacrifice enough memory for standard 1140 crashkernel reservation and where less reliable and 1141 possibly incomplete kdump is preferable to no kdump at 1142 all. 1143 1144 cryptomgr.notests 1145 [KNL] Disable crypto self-tests 1146 1147 cs89x0_dma= [HW,NET] 1148 Format: <dma> 1149 1150 cs89x0_media= [HW,NET] 1151 Format: { rj45 | aui | bnc } 1152 1153 csdlock_debug= [KNL] Enable or disable debug add-ons of cross-CPU 1154 function call handling. When switched on, 1155 additional debug data is printed to the console 1156 in case a hanging CPU is detected, and that 1157 CPU is pinged again in order to try to resolve 1158 the hang situation. The default value of this 1159 option depends on the CSD_LOCK_WAIT_DEBUG_DEFAULT 1160 Kconfig option. 1161 1162 dasd= [HW,NET] 1163 See header of drivers/s390/block/dasd_devmap.c. 1164 1165 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port 1166 (one device per port) 1167 Format: <port#>,<type> 1168 See also Documentation/input/devices/joystick-parport.rst 1169 1170 debug [KNL,EARLY] Enable kernel debugging (events log level). 1171 1172 debug_boot_weak_hash 1173 [KNL,EARLY] Enable printing [hashed] pointers early in the 1174 boot sequence. If enabled, we use a weak hash instead 1175 of siphash to hash pointers. Use this option if you are 1176 seeing instances of '(___ptrval___)') and need to see a 1177 value (hashed pointer) instead. Cryptographically 1178 insecure, please do not use on production kernels. 1179 1180 debug_locks_verbose= 1181 [KNL] verbose locking self-tests 1182 Format: <int> 1183 Print debugging info while doing the locking API 1184 self-tests. 1185 Bitmask for the various LOCKTYPE_ tests. Defaults to 0 1186 (no extra messages), setting it to -1 (all bits set) 1187 will print _a_lot_ more information - normally only 1188 useful to lockdep developers. 1189 1190 debug_objects [KNL,EARLY] Enable object debugging 1191 1192 debug_guardpage_minorder= 1193 [KNL,EARLY] When CONFIG_DEBUG_PAGEALLOC is set, this 1194 parameter allows control of the order of pages that will 1195 be intentionally kept free (and hence protected) by the 1196 buddy allocator. Bigger value increase the probability 1197 of catching random memory corruption, but reduce the 1198 amount of memory for normal system use. The maximum 1199 possible value is MAX_PAGE_ORDER/2. Setting this 1200 parameter to 1 or 2 should be enough to identify most 1201 random memory corruption problems caused by bugs in 1202 kernel or driver code when a CPU writes to (or reads 1203 from) a random memory location. Note that there exists 1204 a class of memory corruptions problems caused by buggy 1205 H/W or F/W or by drivers badly programming DMA 1206 (basically when memory is written at bus level and the 1207 CPU MMU is bypassed) which are not detectable by 1208 CONFIG_DEBUG_PAGEALLOC, hence this option will not 1209 help tracking down these problems. 1210 1211 debug_pagealloc= 1212 [KNL,EARLY] When CONFIG_DEBUG_PAGEALLOC is set, this parameter 1213 enables the feature at boot time. By default, it is 1214 disabled and the system will work mostly the same as a 1215 kernel built without CONFIG_DEBUG_PAGEALLOC. 1216 Note: to get most of debug_pagealloc error reports, it's 1217 useful to also enable the page_owner functionality. 1218 on: enable the feature 1219 1220 debugfs= [KNL,EARLY] This parameter enables what is exposed to 1221 userspace and debugfs internal clients. 1222 Format: { on, off } 1223 on: All functions are enabled. 1224 off: Filesystem is not registered and clients 1225 get a -EPERM as result when trying to register files 1226 or directories within debugfs. 1227 This is equivalent of the runtime functionality if 1228 debugfs was not enabled in the kernel at all. 1229 Default value is set in build-time with a kernel configuration. 1230 1231 debugpat [X86] Enable PAT debugging 1232 1233 default_hugepagesz= 1234 [HW] The size of the default HugeTLB page. This is 1235 the size represented by the legacy /proc/ hugepages 1236 APIs. In addition, this is the default hugetlb size 1237 used for shmget(), mmap() and mounting hugetlbfs 1238 filesystems. If not specified, defaults to the 1239 architecture's default huge page size. Huge page 1240 sizes are architecture dependent. See also 1241 Documentation/admin-guide/mm/hugetlbpage.rst. 1242 Format: size[KMG] 1243 1244 deferred_probe_timeout= 1245 [KNL] Debugging option to set a timeout in seconds for 1246 deferred probe to give up waiting on dependencies to 1247 probe. Only specific dependencies (subsystems or 1248 drivers) that have opted in will be ignored. A timeout 1249 of 0 will timeout at the end of initcalls. If the time 1250 out hasn't expired, it'll be restarted by each 1251 successful driver registration. This option will also 1252 dump out devices still on the deferred probe list after 1253 retrying. 1254 1255 delayacct [KNL] Enable per-task delay accounting 1256 1257 dell_smm_hwmon.ignore_dmi= 1258 [HW] Continue probing hardware even if DMI data 1259 indicates that the driver is running on unsupported 1260 hardware. 1261 1262 dell_smm_hwmon.force= 1263 [HW] Activate driver even if SMM BIOS signature does 1264 not match list of supported models and enable otherwise 1265 blacklisted features. 1266 1267 dell_smm_hwmon.power_status= 1268 [HW] Report power status in /proc/i8k 1269 (disabled by default). 1270 1271 dell_smm_hwmon.restricted= 1272 [HW] Allow controlling fans only if SYS_ADMIN 1273 capability is set. 1274 1275 dell_smm_hwmon.fan_mult= 1276 [HW] Factor to multiply fan speed with. 1277 1278 dell_smm_hwmon.fan_max= 1279 [HW] Maximum configurable fan speed. 1280 1281 dfltcc= [HW,S390] 1282 Format: { on | off | def_only | inf_only | always } 1283 on: s390 zlib hardware support for compression on 1284 level 1 and decompression (default) 1285 off: No s390 zlib hardware support 1286 def_only: s390 zlib hardware support for deflate 1287 only (compression on level 1) 1288 inf_only: s390 zlib hardware support for inflate 1289 only (decompression) 1290 always: Same as 'on' but ignores the selected compression 1291 level always using hardware support (used for debugging) 1292 1293 dhash_entries= [KNL] 1294 Set number of hash buckets for dentry cache. 1295 1296 disable_1tb_segments [PPC,EARLY] 1297 Disables the use of 1TB hash page table segments. This 1298 causes the kernel to fall back to 256MB segments which 1299 can be useful when debugging issues that require an SLB 1300 miss to occur. 1301 1302 disable= [IPV6] 1303 See Documentation/networking/ipv6.rst. 1304 1305 disable_radix [PPC,EARLY] 1306 Disable RADIX MMU mode on POWER9 1307 1308 disable_tlbie [PPC] 1309 Disable TLBIE instruction. Currently does not work 1310 with KVM, with HASH MMU, or with coherent accelerators. 1311 1312 disable_ddw [PPC/PSERIES,EARLY] 1313 Disable Dynamic DMA Window support. Use this 1314 to workaround buggy firmware. 1315 1316 disable_ipv6= [IPV6] 1317 See Documentation/networking/ipv6.rst. 1318 1319 disable_mtrr_cleanup [X86,EARLY] 1320 The kernel tries to adjust MTRR layout from continuous 1321 to discrete, to make X server driver able to add WB 1322 entry later. This parameter disables that. 1323 1324 disable_mtrr_trim [X86, Intel and AMD only,EARLY] 1325 By default the kernel will trim any uncacheable 1326 memory out of your available memory pool based on 1327 MTRR settings. This parameter disables that behavior, 1328 possibly causing your machine to run very slowly. 1329 1330 disable_timer_pin_1 [X86,EARLY] 1331 Disable PIN 1 of APIC timer 1332 Can be useful to work around chipset bugs. 1333 1334 dis_ucode_ldr [X86] Disable the microcode loader. 1335 1336 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support, 1337 this option disables the debugging code at boot. 1338 1339 dma_debug_entries=<number> 1340 This option allows to tune the number of preallocated 1341 entries for DMA-API debugging code. One entry is 1342 required per DMA-API allocation. Use this if the 1343 DMA-API debugging code disables itself because the 1344 architectural default is too low. 1345 1346 dma_debug_driver=<driver_name> 1347 With this option the DMA-API debugging driver 1348 filter feature can be enabled at boot time. Just 1349 pass the driver to filter for as the parameter. 1350 The filter can be disabled or changed to another 1351 driver later using sysfs. 1352 1353 reg_file_data_sampling= 1354 [X86] Controls mitigation for Register File Data 1355 Sampling (RFDS) vulnerability. RFDS is a CPU 1356 vulnerability which may allow userspace to infer 1357 kernel data values previously stored in floating point 1358 registers, vector registers, or integer registers. 1359 RFDS only affects Intel Atom processors. 1360 1361 on: Turns ON the mitigation. 1362 off: Turns OFF the mitigation. 1363 1364 This parameter overrides the compile time default set 1365 by CONFIG_MITIGATION_RFDS. Mitigation cannot be 1366 disabled when other VERW based mitigations (like MDS) 1367 are enabled. In order to disable RFDS mitigation all 1368 VERW based mitigations need to be disabled. 1369 1370 For details see: 1371 Documentation/admin-guide/hw-vuln/reg-file-data-sampling.rst 1372 1373 driver_async_probe= [KNL] 1374 List of driver names to be probed asynchronously. * 1375 matches with all driver names. If * is specified, the 1376 rest of the listed driver names are those that will NOT 1377 match the *. 1378 Format: <driver_name1>,<driver_name2>... 1379 1380 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>] 1381 Broken monitors, graphic adapters, KVMs and EDIDless 1382 panels may send no or incorrect EDID data sets. 1383 This parameter allows to specify an EDID data sets 1384 in the /lib/firmware directory that are used instead. 1385 An EDID data set will only be used for a particular 1386 connector, if its name and a colon are prepended to 1387 the EDID name. Each connector may use a unique EDID 1388 data set by separating the files with a comma. An EDID 1389 data set with no connector name will be used for 1390 any connectors not explicitly specified. 1391 1392 dscc4.setup= [NET] 1393 1394 dt_cpu_ftrs= [PPC,EARLY] 1395 Format: {"off" | "known"} 1396 Control how the dt_cpu_ftrs device-tree binding is 1397 used for CPU feature discovery and setup (if it 1398 exists). 1399 off: Do not use it, fall back to legacy cpu table. 1400 known: Do not pass through unknown features to guests 1401 or userspace, only those that the kernel is aware of. 1402 1403 dump_apple_properties [X86] 1404 Dump name and content of EFI device properties on 1405 x86 Macs. Useful for driver authors to determine 1406 what data is available or for reverse-engineering. 1407 1408 dyndbg[="val"] [KNL,DYNAMIC_DEBUG] 1409 <module>.dyndbg[="val"] 1410 Enable debug messages at boot time. See 1411 Documentation/admin-guide/dynamic-debug-howto.rst 1412 for details. 1413 1414 early_ioremap_debug [KNL,EARLY] 1415 Enable debug messages in early_ioremap support. This 1416 is useful for tracking down temporary early mappings 1417 which are not unmapped. 1418 1419 earlycon= [KNL,EARLY] Output early console device and options. 1420 1421 When used with no options, the early console is 1422 determined by stdout-path property in device tree's 1423 chosen node or the ACPI SPCR table if supported by 1424 the platform. 1425 1426 cdns,<addr>[,options] 1427 Start an early, polled-mode console on a Cadence 1428 (xuartps) serial port at the specified address. Only 1429 supported option is baud rate. If baud rate is not 1430 specified, the serial port must already be setup and 1431 configured. 1432 1433 uart[8250],io,<addr>[,options[,uartclk]] 1434 uart[8250],mmio,<addr>[,options[,uartclk]] 1435 uart[8250],mmio32,<addr>[,options[,uartclk]] 1436 uart[8250],mmio32be,<addr>[,options[,uartclk]] 1437 uart[8250],0x<addr>[,options] 1438 Start an early, polled-mode console on the 8250/16550 1439 UART at the specified I/O port or MMIO address. 1440 MMIO inter-register address stride is either 8-bit 1441 (mmio) or 32-bit (mmio32 or mmio32be). 1442 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed 1443 to be equivalent to 'mmio'. 'options' are specified 1444 in the same format described for "console=ttyS<n>"; if 1445 unspecified, the h/w is not initialized. 'uartclk' is 1446 the uart clock frequency; if unspecified, it is set 1447 to 'BASE_BAUD' * 16. 1448 1449 pl011,<addr> 1450 pl011,mmio32,<addr> 1451 Start an early, polled-mode console on a pl011 serial 1452 port at the specified address. The pl011 serial port 1453 must already be setup and configured. Options are not 1454 yet supported. If 'mmio32' is specified, then only 1455 the driver will use only 32-bit accessors to read/write 1456 the device registers. 1457 1458 liteuart,<addr> 1459 Start an early console on a litex serial port at the 1460 specified address. The serial port must already be 1461 setup and configured. Options are not yet supported. 1462 1463 meson,<addr> 1464 Start an early, polled-mode console on a meson serial 1465 port at the specified address. The serial port must 1466 already be setup and configured. Options are not yet 1467 supported. 1468 1469 msm_serial,<addr> 1470 Start an early, polled-mode console on an msm serial 1471 port at the specified address. The serial port 1472 must already be setup and configured. Options are not 1473 yet supported. 1474 1475 msm_serial_dm,<addr> 1476 Start an early, polled-mode console on an msm serial 1477 dm port at the specified address. The serial port 1478 must already be setup and configured. Options are not 1479 yet supported. 1480 1481 owl,<addr> 1482 Start an early, polled-mode console on a serial port 1483 of an Actions Semi SoC, such as S500 or S900, at the 1484 specified address. The serial port must already be 1485 setup and configured. Options are not yet supported. 1486 1487 rda,<addr> 1488 Start an early, polled-mode console on a serial port 1489 of an RDA Micro SoC, such as RDA8810PL, at the 1490 specified address. The serial port must already be 1491 setup and configured. Options are not yet supported. 1492 1493 sbi 1494 Use RISC-V SBI (Supervisor Binary Interface) for early 1495 console. 1496 1497 smh Use ARM semihosting calls for early console. 1498 1499 s3c2410,<addr> 1500 s3c2412,<addr> 1501 s3c2440,<addr> 1502 s3c6400,<addr> 1503 s5pv210,<addr> 1504 exynos4210,<addr> 1505 Use early console provided by serial driver available 1506 on Samsung SoCs, requires selecting proper type and 1507 a correct base address of the selected UART port. The 1508 serial port must already be setup and configured. 1509 Options are not yet supported. 1510 1511 lantiq,<addr> 1512 Start an early, polled-mode console on a lantiq serial 1513 (lqasc) port at the specified address. The serial port 1514 must already be setup and configured. Options are not 1515 yet supported. 1516 1517 lpuart,<addr> 1518 lpuart32,<addr> 1519 Use early console provided by Freescale LP UART driver 1520 found on Freescale Vybrid and QorIQ LS1021A processors. 1521 A valid base address must be provided, and the serial 1522 port must already be setup and configured. 1523 1524 ec_imx21,<addr> 1525 ec_imx6q,<addr> 1526 Start an early, polled-mode, output-only console on the 1527 Freescale i.MX UART at the specified address. The UART 1528 must already be setup and configured. 1529 1530 ar3700_uart,<addr> 1531 Start an early, polled-mode console on the 1532 Armada 3700 serial port at the specified 1533 address. The serial port must already be setup 1534 and configured. Options are not yet supported. 1535 1536 qcom_geni,<addr> 1537 Start an early, polled-mode console on a Qualcomm 1538 Generic Interface (GENI) based serial port at the 1539 specified address. The serial port must already be 1540 setup and configured. Options are not yet supported. 1541 1542 efifb,[options] 1543 Start an early, unaccelerated console on the EFI 1544 memory mapped framebuffer (if available). On cache 1545 coherent non-x86 systems that use system memory for 1546 the framebuffer, pass the 'ram' option so that it is 1547 mapped with the correct attributes. 1548 1549 linflex,<addr> 1550 Use early console provided by Freescale LINFlexD UART 1551 serial driver for NXP S32V234 SoCs. A valid base 1552 address must be provided, and the serial port must 1553 already be setup and configured. 1554 1555 earlyprintk= [X86,SH,ARM,M68k,S390,UM,EARLY] 1556 earlyprintk=vga 1557 earlyprintk=sclp 1558 earlyprintk=xen 1559 earlyprintk=serial[,ttySn[,baudrate]] 1560 earlyprintk=serial[,0x...[,baudrate]] 1561 earlyprintk=ttySn[,baudrate] 1562 earlyprintk=dbgp[debugController#] 1563 earlyprintk=mmio32,membase[,{nocfg|baudrate}] 1564 earlyprintk=pciserial[,force],bus:device.function[,{nocfg|baudrate}] 1565 earlyprintk=xdbc[xhciController#] 1566 earlyprintk=bios 1567 1568 earlyprintk is useful when the kernel crashes before 1569 the normal console is initialized. It is not enabled by 1570 default because it has some cosmetic problems. 1571 1572 Use "nocfg" to skip UART configuration, assume 1573 BIOS/firmware has configured UART correctly. 1574 1575 Append ",keep" to not disable it when the real console 1576 takes over. 1577 1578 Only one of vga, serial, or usb debug port can 1579 be used at a time. 1580 1581 Currently only ttyS0 and ttyS1 may be specified by 1582 name. Other I/O ports may be explicitly specified 1583 on some architectures (x86 and arm at least) by 1584 replacing ttySn with an I/O port address, like this: 1585 earlyprintk=serial,0x1008,115200 1586 You can find the port for a given device in 1587 /proc/tty/driver/serial: 1588 2: uart:ST16650V2 port:00001008 irq:18 ... 1589 1590 Interaction with the standard serial driver is not 1591 very good. 1592 1593 The VGA output is eventually overwritten by 1594 the real console. 1595 1596 The xen option can only be used in Xen domains. 1597 1598 The sclp output can only be used on s390. 1599 1600 The bios output can only be used on SuperH. 1601 1602 The optional "force" to "pciserial" enables use of a 1603 PCI device even when its classcode is not of the 1604 UART class. 1605 1606 edac_report= [HW,EDAC] Control how to report EDAC event 1607 Format: {"on" | "off" | "force"} 1608 on: enable EDAC to report H/W event. May be overridden 1609 by other higher priority error reporting module. 1610 off: disable H/W event reporting through EDAC. 1611 force: enforce the use of EDAC to report H/W event. 1612 default: on. 1613 1614 edd= [EDD] 1615 Format: {"off" | "on" | "skip[mbr]"} 1616 1617 efi= [EFI,EARLY] 1618 Format: { "debug", "disable_early_pci_dma", 1619 "nochunk", "noruntime", "nosoftreserve", 1620 "novamap", "no_disable_early_pci_dma" } 1621 debug: enable misc debug output. 1622 disable_early_pci_dma: disable the busmaster bit on all 1623 PCI bridges while in the EFI boot stub. 1624 nochunk: disable reading files in "chunks" in the EFI 1625 boot stub, as chunking can cause problems with some 1626 firmware implementations. 1627 noruntime : disable EFI runtime services support 1628 nosoftreserve: The EFI_MEMORY_SP (Specific Purpose) 1629 attribute may cause the kernel to reserve the 1630 memory range for a memory mapping driver to 1631 claim. Specify efi=nosoftreserve to disable this 1632 reservation and treat the memory by its base type 1633 (i.e. EFI_CONVENTIONAL_MEMORY / "System RAM"). 1634 novamap: do not call SetVirtualAddressMap(). 1635 no_disable_early_pci_dma: Leave the busmaster bit set 1636 on all PCI bridges while in the EFI boot stub 1637 1638 efi_no_storage_paranoia [EFI,X86,EARLY] 1639 Using this parameter you can use more than 50% of 1640 your efi variable storage. Use this parameter only if 1641 you are really sure that your UEFI does sane gc and 1642 fulfills the spec otherwise your board may brick. 1643 1644 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT 1645 that is to be dynamically loaded by Linux. If there are 1646 multiple variables with the same name but with different 1647 vendor GUIDs, all of them will be loaded. See 1648 Documentation/admin-guide/acpi/ssdt-overlays.rst for details. 1649 1650 1651 eisa_irq_edge= [PARISC,HW] 1652 See header of drivers/parisc/eisa.c. 1653 1654 ekgdboc= [X86,KGDB,EARLY] Allow early kernel console debugging 1655 Format: ekgdboc=kbd 1656 1657 This is designed to be used in conjunction with 1658 the boot argument: earlyprintk=vga 1659 1660 This parameter works in place of the kgdboc parameter 1661 but can only be used if the backing tty is available 1662 very early in the boot process. For early debugging 1663 via a serial port see kgdboc_earlycon instead. 1664 1665 elanfreq= [X86-32] 1666 See comment before function elanfreq_setup() in 1667 arch/x86/kernel/cpu/cpufreq/elanfreq.c. 1668 1669 elfcorehdr=[size[KMG]@]offset[KMG] [PPC,SH,X86,S390,EARLY] 1670 Specifies physical address of start of kernel core 1671 image elf header and optionally the size. Generally 1672 kexec loader will pass this option to capture kernel. 1673 See Documentation/admin-guide/kdump/kdump.rst for details. 1674 1675 enable_mtrr_cleanup [X86,EARLY] 1676 The kernel tries to adjust MTRR layout from continuous 1677 to discrete, to make X server driver able to add WB 1678 entry later. This parameter enables that. 1679 1680 enable_timer_pin_1 [X86] 1681 Enable PIN 1 of APIC timer 1682 Can be useful to work around chipset bugs 1683 (in particular on some ATI chipsets). 1684 The kernel tries to set a reasonable default. 1685 1686 enforcing= [SELINUX] Set initial enforcing status. 1687 Format: {"0" | "1"} 1688 See security/selinux/Kconfig help text. 1689 0 -- permissive (log only, no denials). 1690 1 -- enforcing (deny and log). 1691 Default value is 0. 1692 Value can be changed at runtime via 1693 /sys/fs/selinux/enforce. 1694 1695 erst_disable [ACPI] 1696 Disable Error Record Serialization Table (ERST) 1697 support. 1698 1699 ether= [HW,NET] Ethernet cards parameters 1700 This option is obsoleted by the "netdev=" option, which 1701 has equivalent usage. See its documentation for details. 1702 1703 evm= [EVM] 1704 Format: { "fix" } 1705 Permit 'security.evm' to be updated regardless of 1706 current integrity status. 1707 1708 early_page_ext [KNL,EARLY] Enforces page_ext initialization to earlier 1709 stages so cover more early boot allocations. 1710 Please note that as side effect some optimizations 1711 might be disabled to achieve that (e.g. parallelized 1712 memory initialization is disabled) so the boot process 1713 might take longer, especially on systems with a lot of 1714 memory. Available with CONFIG_PAGE_EXTENSION=y. 1715 1716 failslab= 1717 fail_usercopy= 1718 fail_page_alloc= 1719 fail_skb_realloc= 1720 fail_make_request=[KNL] 1721 General fault injection mechanism. 1722 Format: <interval>,<probability>,<space>,<times> 1723 See also Documentation/fault-injection/. 1724 1725 fb_tunnels= [NET] 1726 Format: { initns | none } 1727 See Documentation/admin-guide/sysctl/net.rst for 1728 fb_tunnels_only_for_init_ns 1729 1730 floppy= [HW] 1731 See Documentation/admin-guide/blockdev/floppy.rst. 1732 1733 forcepae [X86-32] 1734 Forcefully enable Physical Address Extension (PAE). 1735 Many Pentium M systems disable PAE but may have a 1736 functionally usable PAE implementation. 1737 Warning: use of this parameter will taint the kernel 1738 and may cause unknown problems. 1739 1740 fred= [X86-64] 1741 Enable/disable Flexible Return and Event Delivery. 1742 Format: { on | off } 1743 on: enable FRED when it's present. 1744 off: disable FRED, the default setting. 1745 1746 ftrace=[tracer] 1747 [FTRACE] will set and start the specified tracer 1748 as early as possible in order to facilitate early 1749 boot debugging. 1750 1751 ftrace_boot_snapshot 1752 [FTRACE] On boot up, a snapshot will be taken of the 1753 ftrace ring buffer that can be read at: 1754 /sys/kernel/tracing/snapshot. 1755 This is useful if you need tracing information from kernel 1756 boot up that is likely to be overridden by user space 1757 start up functionality. 1758 1759 Optionally, the snapshot can also be defined for a tracing 1760 instance that was created by the trace_instance= command 1761 line parameter. 1762 1763 trace_instance=foo,sched_switch ftrace_boot_snapshot=foo 1764 1765 The above will cause the "foo" tracing instance to trigger 1766 a snapshot at the end of boot up. 1767 1768 ftrace_dump_on_oops[=2(orig_cpu) | =<instance>][,<instance> | 1769 ,<instance>=2(orig_cpu)] 1770 [FTRACE] will dump the trace buffers on oops. 1771 If no parameter is passed, ftrace will dump global 1772 buffers of all CPUs, if you pass 2 or orig_cpu, it 1773 will dump only the buffer of the CPU that triggered 1774 the oops, or the specific instance will be dumped if 1775 its name is passed. Multiple instance dump is also 1776 supported, and instances are separated by commas. Each 1777 instance supports only dump on CPU that triggered the 1778 oops by passing 2 or orig_cpu to it. 1779 1780 ftrace_dump_on_oops=foo=orig_cpu 1781 1782 The above will dump only the buffer of "foo" instance 1783 on CPU that triggered the oops. 1784 1785 ftrace_dump_on_oops,foo,bar=orig_cpu 1786 1787 The above will dump global buffer on all CPUs, the 1788 buffer of "foo" instance on all CPUs and the buffer 1789 of "bar" instance on CPU that triggered the oops. 1790 1791 ftrace_filter=[function-list] 1792 [FTRACE] Limit the functions traced by the function 1793 tracer at boot up. function-list is a comma-separated 1794 list of functions. This list can be changed at run 1795 time by the set_ftrace_filter file in the debugfs 1796 tracing directory. 1797 1798 ftrace_notrace=[function-list] 1799 [FTRACE] Do not trace the functions specified in 1800 function-list. This list can be changed at run time 1801 by the set_ftrace_notrace file in the debugfs 1802 tracing directory. 1803 1804 ftrace_graph_filter=[function-list] 1805 [FTRACE] Limit the top level callers functions traced 1806 by the function graph tracer at boot up. 1807 function-list is a comma-separated list of functions 1808 that can be changed at run time by the 1809 set_graph_function file in the debugfs tracing directory. 1810 1811 ftrace_graph_notrace=[function-list] 1812 [FTRACE] Do not trace from the functions specified in 1813 function-list. This list is a comma-separated list of 1814 functions that can be changed at run time by the 1815 set_graph_notrace file in the debugfs tracing directory. 1816 1817 ftrace_graph_max_depth=<uint> 1818 [FTRACE] Used with the function graph tracer. This is 1819 the max depth it will trace into a function. This value 1820 can be changed at run time by the max_graph_depth file 1821 in the tracefs tracing directory. default: 0 (no limit) 1822 1823 fw_devlink= [KNL,EARLY] Create device links between consumer and supplier 1824 devices by scanning the firmware to infer the 1825 consumer/supplier relationships. This feature is 1826 especially useful when drivers are loaded as modules as 1827 it ensures proper ordering of tasks like device probing 1828 (suppliers first, then consumers), supplier boot state 1829 clean up (only after all consumers have probed), 1830 suspend/resume & runtime PM (consumers first, then 1831 suppliers). 1832 Format: { off | permissive | on | rpm } 1833 off -- Don't create device links from firmware info. 1834 permissive -- Create device links from firmware info 1835 but use it only for ordering boot state clean 1836 up (sync_state() calls). 1837 on -- Create device links from firmware info and use it 1838 to enforce probe and suspend/resume ordering. 1839 rpm -- Like "on", but also use to order runtime PM. 1840 1841 fw_devlink.strict=<bool> 1842 [KNL,EARLY] Treat all inferred dependencies as mandatory 1843 dependencies. This only applies for fw_devlink=on|rpm. 1844 Format: <bool> 1845 1846 fw_devlink.sync_state = 1847 [KNL,EARLY] When all devices that could probe have finished 1848 probing, this parameter controls what to do with 1849 devices that haven't yet received their sync_state() 1850 calls. 1851 Format: { strict | timeout } 1852 strict -- Default. Continue waiting on consumers to 1853 probe successfully. 1854 timeout -- Give up waiting on consumers and call 1855 sync_state() on any devices that haven't yet 1856 received their sync_state() calls after 1857 deferred_probe_timeout has expired or by 1858 late_initcall() if !CONFIG_MODULES. 1859 1860 gamecon.map[2|3]= 1861 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad 1862 support via parallel port (up to 5 devices per port) 1863 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5> 1864 See also Documentation/input/devices/joystick-parport.rst 1865 1866 gamma= [HW,DRM] 1867 1868 gart_fix_e820= [X86-64,EARLY] disable the fix e820 for K8 GART 1869 Format: off | on 1870 default: on 1871 1872 gather_data_sampling= 1873 [X86,INTEL,EARLY] Control the Gather Data Sampling (GDS) 1874 mitigation. 1875 1876 Gather Data Sampling is a hardware vulnerability which 1877 allows unprivileged speculative access to data which was 1878 previously stored in vector registers. 1879 1880 This issue is mitigated by default in updated microcode. 1881 The mitigation may have a performance impact but can be 1882 disabled. On systems without the microcode mitigation 1883 disabling AVX serves as a mitigation. 1884 1885 force: Disable AVX to mitigate systems without 1886 microcode mitigation. No effect if the microcode 1887 mitigation is present. Known to cause crashes in 1888 userspace with buggy AVX enumeration. 1889 1890 off: Disable GDS mitigation. 1891 1892 gbpages [X86] Use GB pages for kernel direct mappings. 1893 1894 gcov_persist= [GCOV] When non-zero (default), profiling data for 1895 kernel modules is saved and remains accessible via 1896 debugfs, even when the module is unloaded/reloaded. 1897 When zero, profiling data is discarded and associated 1898 debugfs files are removed at module unload time. 1899 1900 goldfish [X86] Enable the goldfish android emulator platform. 1901 Don't use this when you are not running on the 1902 android emulator 1903 1904 gpio-mockup.gpio_mockup_ranges 1905 [HW] Sets the ranges of gpiochip of for this device. 1906 Format: <start1>,<end1>,<start2>,<end2>... 1907 gpio-mockup.gpio_mockup_named_lines 1908 [HW] Let the driver know GPIO lines should be named. 1909 1910 gpt [EFI] Forces disk with valid GPT signature but 1911 invalid Protective MBR to be treated as GPT. If the 1912 primary GPT is corrupted, it enables the backup/alternate 1913 GPT to be used instead. 1914 1915 grcan.enable0= [HW] Configuration of physical interface 0. Determines 1916 the "Enable 0" bit of the configuration register. 1917 Format: 0 | 1 1918 Default: 0 1919 grcan.enable1= [HW] Configuration of physical interface 1. Determines 1920 the "Enable 0" bit of the configuration register. 1921 Format: 0 | 1 1922 Default: 0 1923 grcan.select= [HW] Select which physical interface to use. 1924 Format: 0 | 1 1925 Default: 0 1926 grcan.txsize= [HW] Sets the size of the tx buffer. 1927 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0. 1928 Default: 1024 1929 grcan.rxsize= [HW] Sets the size of the rx buffer. 1930 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0. 1931 Default: 1024 1932 1933 hardened_usercopy= 1934 [KNL] Under CONFIG_HARDENED_USERCOPY, whether 1935 hardening is enabled for this boot. Hardened 1936 usercopy checking is used to protect the kernel 1937 from reading or writing beyond known memory 1938 allocation boundaries as a proactive defense 1939 against bounds-checking flaws in the kernel's 1940 copy_to_user()/copy_from_user() interface. 1941 The default is determined by 1942 CONFIG_HARDENED_USERCOPY_DEFAULT_ON. 1943 on Perform hardened usercopy checks. 1944 off Disable hardened usercopy checks. 1945 1946 hardlockup_all_cpu_backtrace= 1947 [KNL] Should the hard-lockup detector generate 1948 backtraces on all cpus. 1949 Format: 0 | 1 1950 1951 hash_pointers= 1952 [KNL,EARLY] 1953 By default, when pointers are printed to the console 1954 or buffers via the %p format string, that pointer is 1955 "hashed", i.e. obscured by hashing the pointer value. 1956 This is a security feature that hides actual kernel 1957 addresses from unprivileged users, but it also makes 1958 debugging the kernel more difficult since unequal 1959 pointers can no longer be compared. The choices are: 1960 Format: { auto | always | never } 1961 Default: auto 1962 1963 auto - Hash pointers unless slab_debug is enabled. 1964 always - Always hash pointers (even if slab_debug is 1965 enabled). 1966 never - Never hash pointers. This option should only 1967 be specified when debugging the kernel. Do 1968 not use on production kernels. The boot 1969 param "no_hash_pointers" is an alias for 1970 this mode. 1971 1972 For controlling hashing dynamically at runtime, 1973 use the "kernel.kptr_restrict" sysctl instead. 1974 1975 hashdist= [KNL,NUMA] Large hashes allocated during boot 1976 are distributed across NUMA nodes. Defaults on 1977 for 64-bit NUMA, off otherwise. 1978 Format: 0 | 1 (for off | on) 1979 1980 hd= [EIDE] (E)IDE hard drive subsystem geometry 1981 Format: <cyl>,<head>,<sect> 1982 1983 hest_disable [ACPI] 1984 Disable Hardware Error Source Table (HEST) support; 1985 corresponding firmware-first mode error processing 1986 logic will be disabled. 1987 1988 hibernate= [HIBERNATION] 1989 noresume Don't check if there's a hibernation image 1990 present during boot. 1991 nocompress Don't compress/decompress hibernation images. 1992 no Disable hibernation and resume. 1993 protect_image Turn on image protection during restoration 1994 (that will set all pages holding image data 1995 during restoration read-only). 1996 1997 hibernate.compressor= [HIBERNATION] Compression algorithm to be 1998 used with hibernation. 1999 Format: { lzo | lz4 } 2000 Default: lzo 2001 2002 lzo: Select LZO compression algorithm to 2003 compress/decompress hibernation image. 2004 2005 lz4: Select LZ4 compression algorithm to 2006 compress/decompress hibernation image. 2007 2008 hibernate.pm_test_delay= 2009 [HIBERNATION] 2010 Sets the number of seconds to remain in a hibernation test 2011 mode before resuming the system (see 2012 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG 2013 is set. Default value is 5. 2014 2015 hibernate_compression_threads= 2016 [HIBERNATION] 2017 Set the number of threads used for compressing or decompressing 2018 hibernation images. 2019 2020 Format: <integer> 2021 Default: 3 2022 Minimum: 1 2023 Example: hibernate_compression_threads=4 2024 2025 highmem=nn[KMG] [KNL,BOOT,EARLY] forces the highmem zone to have an exact 2026 size of <nn>. This works even on boxes that have no 2027 highmem otherwise. This also works to reduce highmem 2028 size on bigger boxes. 2029 2030 highres= [KNL] Enable/disable high resolution timer mode. 2031 Valid parameters: "on", "off" 2032 Default: "on" 2033 2034 hlt [BUGS=ARM,SH] 2035 2036 hostname= [KNL,EARLY] Set the hostname (aka UTS nodename). 2037 Format: <string> 2038 This allows setting the system's hostname during early 2039 startup. This sets the name returned by gethostname. 2040 Using this parameter to set the hostname makes it 2041 possible to ensure the hostname is correctly set before 2042 any userspace processes run, avoiding the possibility 2043 that a process may call gethostname before the hostname 2044 has been explicitly set, resulting in the calling 2045 process getting an incorrect result. The string must 2046 not exceed the maximum allowed hostname length (usually 2047 64 characters) and will be truncated otherwise. 2048 2049 hpet= [X86-32,HPET] option to control HPET usage 2050 Format: { enable (default) | disable | force | 2051 verbose } 2052 disable: disable HPET and use PIT instead 2053 force: allow force enabled of undocumented chips (ICH4, 2054 VIA, nVidia) 2055 verbose: show contents of HPET registers during setup 2056 2057 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET 2058 registers. Default set by CONFIG_HPET_MMAP_DEFAULT. 2059 2060 hugepages= [HW,EARLY] Number of HugeTLB pages to allocate at boot. 2061 If this follows hugepagesz (below), it specifies 2062 the number of pages of hugepagesz to be allocated. 2063 If this is the first HugeTLB parameter on the command 2064 line, it specifies the number of pages to allocate for 2065 the default huge page size. If using node format, the 2066 number of pages to allocate per-node can be specified. 2067 See also Documentation/admin-guide/mm/hugetlbpage.rst. 2068 Format: <integer> or (node format) 2069 <node>:<integer>[,<node>:<integer>] 2070 2071 hugepagesz= 2072 [HW,EARLY] The size of the HugeTLB pages. This is 2073 used in conjunction with hugepages (above) to 2074 allocate huge pages of a specific size at boot. The 2075 pair hugepagesz=X hugepages=Y can be specified once 2076 for each supported huge page size. Huge page sizes 2077 are architecture dependent. See also 2078 Documentation/admin-guide/mm/hugetlbpage.rst. 2079 Format: size[KMG] 2080 2081 hugepage_alloc_threads= 2082 [HW] The number of threads that should be used to 2083 allocate hugepages during boot. This option can be 2084 used to improve system bootup time when allocating 2085 a large amount of huge pages. 2086 The default value is 25% of the available hardware threads. 2087 2088 Note that this parameter only applies to non-gigantic huge pages. 2089 2090 hugetlb_cma= [HW,CMA,EARLY] The size of a CMA area used for allocation 2091 of gigantic hugepages. Or using node format, the size 2092 of a CMA area per node can be specified. 2093 Format: nn[KMGTPE] or (node format) 2094 <node>:nn[KMGTPE][,<node>:nn[KMGTPE]] 2095 2096 Reserve a CMA area of given size and allocate gigantic 2097 hugepages using the CMA allocator. If enabled, the 2098 boot-time allocation of gigantic hugepages is skipped. 2099 2100 hugetlb_cma_only= 2101 [HW,CMA,EARLY] When allocating new HugeTLB pages, only 2102 try to allocate from the CMA areas. 2103 2104 This option does nothing if hugetlb_cma= is not also 2105 specified. 2106 2107 hugetlb_free_vmemmap= 2108 [KNL] Requires CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP 2109 enabled. 2110 Control if HugeTLB Vmemmap Optimization (HVO) is enabled. 2111 Allows heavy hugetlb users to free up some more 2112 memory (7 * PAGE_SIZE for each 2MB hugetlb page). 2113 Format: { on | off (default) } 2114 2115 on: enable HVO 2116 off: disable HVO 2117 2118 Built with CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP_DEFAULT_ON=y, 2119 the default is on. 2120 2121 Note that the vmemmap pages may be allocated from the added 2122 memory block itself when memory_hotplug.memmap_on_memory is 2123 enabled, those vmemmap pages cannot be optimized even if this 2124 feature is enabled. Other vmemmap pages not allocated from 2125 the added memory block itself do not be affected. 2126 2127 hung_task_panic= 2128 [KNL] Number of hung tasks to trigger kernel panic. 2129 Format: <int> 2130 2131 When set to a non-zero value, a kernel panic will be triggered if 2132 the number of detected hung tasks reaches this value. 2133 2134 0: don't panic 2135 1: panic immediately on first hung task 2136 N: panic after N hung tasks are detected in a single scan 2137 2138 The default value is controlled by the 2139 CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time option. The value 2140 selected by this boot parameter can be changed later by the 2141 kernel.hung_task_panic sysctl. 2142 2143 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC) 2144 terminal devices. Valid values: 0..8 2145 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs. 2146 If specified, z/VM IUCV HVC accepts connections 2147 from listed z/VM user IDs only. 2148 2149 hv_nopvspin [X86,HYPER_V,EARLY] 2150 Disables the paravirt spinlock optimizations 2151 which allow the hypervisor to 'idle' the guest 2152 on lock contention. 2153 2154 hw_protection= [HW] 2155 Format: reboot | shutdown 2156 2157 Hardware protection action taken on critical events like 2158 overtemperature or imminent voltage loss. 2159 2160 i2c_bus= [HW] Override the default board specific I2C bus speed 2161 or register an additional I2C bus that is not 2162 registered from board initialization code. 2163 Format: 2164 <bus_id>,<clkrate> 2165 2166 i2c_touchscreen_props= [HW,ACPI,X86] 2167 Set device-properties for ACPI-enumerated I2C-attached 2168 touchscreen, to e.g. fix coordinates of upside-down 2169 mounted touchscreens. If you need this option please 2170 submit a drivers/platform/x86/touchscreen_dmi.c patch 2171 adding a DMI quirk for this. 2172 2173 Format: 2174 <ACPI_HW_ID>:<prop_name>=<val>[:prop_name=val][:...] 2175 Where <val> is one of: 2176 Omit "=<val>" entirely Set a boolean device-property 2177 Unsigned number Set a u32 device-property 2178 Anything else Set a string device-property 2179 2180 Examples (split over multiple lines): 2181 i2c_touchscreen_props=GDIX1001:touchscreen-inverted-x: 2182 touchscreen-inverted-y 2183 2184 i2c_touchscreen_props=MSSL1680:touchscreen-size-x=1920: 2185 touchscreen-size-y=1080:touchscreen-inverted-y: 2186 firmware-name=gsl1680-vendor-model.fw:silead,home-button 2187 2188 i8042.debug [HW] Toggle i8042 debug mode 2189 i8042.unmask_kbd_data 2190 [HW] Enable printing of interrupt data from the KBD port 2191 (disabled by default, and as a pre-condition 2192 requires that i8042.debug=1 be enabled) 2193 i8042.direct [HW] Put keyboard port into non-translated mode 2194 i8042.dumbkbd [HW] Pretend that controller can only read data from 2195 keyboard and cannot control its state 2196 (Don't attempt to blink the leds) 2197 i8042.noaux [HW] Don't check for auxiliary (== mouse) port 2198 i8042.nokbd [HW] Don't check/create keyboard port 2199 i8042.noloop [HW] Disable the AUX Loopback command while probing 2200 for the AUX port 2201 i8042.nomux [HW] Don't check presence of an active multiplexing 2202 controller 2203 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX 2204 controllers 2205 i8042.notimeout [HW] Ignore timeout condition signalled by controller 2206 i8042.reset [HW] Reset the controller during init, cleanup and 2207 suspend-to-ram transitions, only during s2r 2208 transitions, or never reset 2209 Format: { 1 | Y | y | 0 | N | n } 2210 1, Y, y: always reset controller 2211 0, N, n: don't ever reset controller 2212 Default: only on s2r transitions on x86; most other 2213 architectures force reset to be always executed 2214 i8042.unlock [HW] Unlock (ignore) the keylock 2215 i8042.kbdreset [HW] Reset device connected to KBD port 2216 i8042.probe_defer 2217 [HW] Allow deferred probing upon i8042 probe errors 2218 2219 i810= [HW,DRM] 2220 2221 i915.invert_brightness= 2222 [DRM] Invert the sense of the variable that is used to 2223 set the brightness of the panel backlight. Normally a 2224 brightness value of 0 indicates backlight switched off, 2225 and the maximum of the brightness value sets the backlight 2226 to maximum brightness. If this parameter is set to 0 2227 (default) and the machine requires it, or this parameter 2228 is set to 1, a brightness value of 0 sets the backlight 2229 to maximum brightness, and the maximum of the brightness 2230 value switches the backlight off. 2231 -1 -- never invert brightness 2232 0 -- machine default 2233 1 -- force brightness inversion 2234 2235 ia32_emulation= [X86-64] 2236 Format: <bool> 2237 When true, allows loading 32-bit programs and executing 32-bit 2238 syscalls, essentially overriding IA32_EMULATION_DEFAULT_DISABLED at 2239 boot time. When false, unconditionally disables IA32 emulation. 2240 2241 icn= [HW,ISDN] 2242 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]] 2243 2244 2245 idle= [X86,EARLY] 2246 Format: idle=poll, idle=halt, idle=nomwait 2247 2248 idle=poll: Don't do power saving in the idle loop 2249 using HLT, but poll for rescheduling event. This will 2250 make the CPUs eat a lot more power, but may be useful 2251 to get slightly better performance in multiprocessor 2252 benchmarks. It also makes some profiling using 2253 performance counters more accurate. Please note that 2254 on systems with MONITOR/MWAIT support (like Intel 2255 EM64T CPUs) this option has no performance advantage 2256 over the normal idle loop. It may also interact badly 2257 with hyperthreading. 2258 2259 idle=halt: Halt is forced to be used for CPU idle. 2260 In such case C2/C3 won't be used again. 2261 2262 idle=nomwait: Disable mwait for CPU C-states 2263 2264 idxd.sva= [HW] 2265 Format: <bool> 2266 Allow force disabling of Shared Virtual Memory (SVA) 2267 support for the idxd driver. By default it is set to 2268 true (1). 2269 2270 idxd.tc_override= [HW] 2271 Format: <bool> 2272 Allow override of default traffic class configuration 2273 for the device. By default it is set to false (0). 2274 2275 ieee754= [MIPS] Select IEEE Std 754 conformance mode 2276 Format: { strict | legacy | 2008 | relaxed | emulated } 2277 Default: strict 2278 2279 Choose which programs will be accepted for execution 2280 based on the IEEE 754 NaN encoding(s) supported by 2281 the FPU and the NaN encoding requested with the value 2282 of an ELF file header flag individually set by each 2283 binary. Hardware implementations are permitted to 2284 support either or both of the legacy and the 2008 NaN 2285 encoding mode. 2286 2287 Available settings are as follows: 2288 strict accept binaries that request a NaN encoding 2289 supported by the FPU 2290 legacy only accept legacy-NaN binaries, if supported 2291 by the FPU 2292 2008 only accept 2008-NaN binaries, if supported 2293 by the FPU 2294 relaxed accept any binaries regardless of whether 2295 supported by the FPU 2296 emulated accept any binaries but enable FPU emulator 2297 if binary mode is unsupported by the FPU. 2298 2299 The FPU emulator is always able to support both NaN 2300 encodings, so if no FPU hardware is present or it has 2301 been disabled with 'nofpu', then the settings of 2302 'legacy' and '2008' strap the emulator accordingly, 2303 'relaxed' straps the emulator for both legacy-NaN and 2304 2008-NaN, whereas 'strict' enables legacy-NaN only on 2305 legacy processors and both NaN encodings on MIPS32 or 2306 MIPS64 CPUs. 2307 2308 The setting for ABS.fmt/NEG.fmt instruction execution 2309 mode generally follows that for the NaN encoding, 2310 except where unsupported by hardware. 2311 2312 ignore_loglevel [KNL,EARLY] 2313 Ignore loglevel setting - this will print /all/ 2314 kernel messages to the console. Useful for debugging. 2315 We also add it as printk module parameter, so users 2316 could change it dynamically, usually by 2317 /sys/module/printk/parameters/ignore_loglevel. 2318 2319 ignore_rlimit_data 2320 Ignore RLIMIT_DATA setting for data mappings, 2321 print warning at first misuse. Can be changed via 2322 /sys/module/kernel/parameters/ignore_rlimit_data. 2323 2324 ihash_entries= [KNL] 2325 Set number of hash buckets for inode cache. 2326 2327 ima_appraise= [IMA] appraise integrity measurements 2328 Format: { "off" | "enforce" | "fix" | "log" } 2329 default: "enforce" 2330 2331 ima_appraise_tcb [IMA] Deprecated. Use ima_policy= instead. 2332 The builtin appraise policy appraises all files 2333 owned by uid=0. 2334 2335 ima_canonical_fmt [IMA] 2336 Use the canonical format for the binary runtime 2337 measurements, instead of host native format. 2338 2339 ima_hash= [IMA] 2340 Format: { md5 | sha1 | rmd160 | sha256 | sha384 2341 | sha512 | ... } 2342 default: "sha1" 2343 2344 The list of supported hash algorithms is defined 2345 in crypto/hash_info.h. 2346 2347 ima_policy= [IMA] 2348 The builtin policies to load during IMA setup. 2349 Format: "tcb | appraise_tcb | secure_boot | 2350 fail_securely | critical_data" 2351 2352 The "tcb" policy measures all programs exec'd, files 2353 mmap'd for exec, and all files opened with the read 2354 mode bit set by either the effective uid (euid=0) or 2355 uid=0. 2356 2357 The "appraise_tcb" policy appraises the integrity of 2358 all files owned by root. 2359 2360 The "secure_boot" policy appraises the integrity 2361 of files (eg. kexec kernel image, kernel modules, 2362 firmware, policy, etc) based on file signatures. 2363 2364 The "fail_securely" policy forces file signature 2365 verification failure also on privileged mounted 2366 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE 2367 flag. 2368 2369 The "critical_data" policy measures kernel integrity 2370 critical data. 2371 2372 ima_tcb [IMA] Deprecated. Use ima_policy= instead. 2373 Load a policy which meets the needs of the Trusted 2374 Computing Base. This means IMA will measure all 2375 programs exec'd, files mmap'd for exec, and all files 2376 opened for read by uid=0. 2377 2378 ima_template= [IMA] 2379 Select one of defined IMA measurements template formats. 2380 Formats: { "ima" | "ima-ng" | "ima-ngv2" | "ima-sig" | 2381 "ima-sigv2" } 2382 Default: "ima-ng" 2383 2384 ima_template_fmt= 2385 [IMA] Define a custom template format. 2386 Format: { "field1|...|fieldN" } 2387 2388 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage 2389 Format: <min_file_size> 2390 Set the minimal file size for using asynchronous hash. 2391 If left unspecified, ahash usage is disabled. 2392 2393 ahash performance varies for different data sizes on 2394 different crypto accelerators. This option can be used 2395 to achieve the best performance for a particular HW. 2396 2397 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size 2398 Format: <bufsize> 2399 Set hashing buffer size. Default: 4k. 2400 2401 ahash performance varies for different chunk sizes on 2402 different crypto accelerators. This option can be used 2403 to achieve best performance for particular HW. 2404 2405 ima= [IMA] Enable or disable IMA 2406 Format: { "off" | "on" } 2407 Default: "on" 2408 Note that disabling IMA is limited to kdump kernel. 2409 2410 indirect_target_selection= [X86,Intel] Mitigation control for Indirect 2411 Target Selection(ITS) bug in Intel CPUs. Updated 2412 microcode is also required for a fix in IBPB. 2413 2414 on: Enable mitigation (default). 2415 off: Disable mitigation. 2416 force: Force the ITS bug and deploy default 2417 mitigation. 2418 vmexit: Only deploy mitigation if CPU is affected by 2419 guest/host isolation part of ITS. 2420 stuff: Deploy RSB-fill mitigation when retpoline is 2421 also deployed. Otherwise, deploy the default 2422 mitigation. 2423 2424 For details see: 2425 Documentation/admin-guide/hw-vuln/indirect-target-selection.rst 2426 2427 init= [KNL] 2428 Format: <full_path> 2429 Run specified binary instead of /sbin/init as init 2430 process. 2431 2432 initcall_debug [KNL] Trace initcalls as they are executed. Useful 2433 for working out where the kernel is dying during 2434 startup. 2435 2436 initcall_blacklist= [KNL] Do not execute a comma-separated list of 2437 initcall functions. Useful for debugging built-in 2438 modules and initcalls. 2439 2440 initramfs_async= [KNL] 2441 Format: <bool> 2442 Default: 1 2443 This parameter controls whether the initramfs 2444 image is unpacked asynchronously, concurrently 2445 with devices being probed and 2446 initialized. This should normally just work, 2447 but as a debugging aid, one can get the 2448 historical behaviour of the initramfs 2449 unpacking being completed before device_ and 2450 late_ initcalls. 2451 2452 initrd= [BOOT,EARLY] Specify the location of the initial ramdisk 2453 2454 initrdmem= [KNL,EARLY] Specify a physical address and size from which to 2455 load the initrd. If an initrd is compiled in or 2456 specified in the bootparams, it takes priority over this 2457 setting. 2458 Format: ss[KMG],nn[KMG] 2459 Default is 0, 0 2460 2461 init_on_alloc= [MM,EARLY] Fill newly allocated pages and heap objects with 2462 zeroes. 2463 Format: 0 | 1 2464 Default set by CONFIG_INIT_ON_ALLOC_DEFAULT_ON. 2465 2466 init_on_free= [MM,EARLY] Fill freed pages and heap objects with zeroes. 2467 Format: 0 | 1 2468 Default set by CONFIG_INIT_ON_FREE_DEFAULT_ON. 2469 2470 init_pkru= [X86] Specify the default memory protection keys rights 2471 register contents for all processes. 0x55555554 by 2472 default (disallow access to all but pkey 0). Can 2473 override in debugfs after boot. 2474 2475 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver 2476 Format: <irq> 2477 2478 int_pln_enable [X86] Enable power limit notification interrupt 2479 2480 integrity_audit=[IMA] 2481 Format: { "0" | "1" } 2482 0 -- basic integrity auditing messages. (Default) 2483 1 -- additional integrity auditing messages. 2484 2485 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option 2486 on 2487 Enable intel iommu driver. 2488 off 2489 Disable intel iommu driver. 2490 igfx_off [Default Off] 2491 By default, gfx is mapped as normal device. If a gfx 2492 device has a dedicated DMAR unit, the DMAR unit is 2493 bypassed by not enabling DMAR with this option. In 2494 this case, gfx device will use physical address for 2495 DMA. 2496 strict [Default Off] 2497 Deprecated, equivalent to iommu.strict=1. 2498 sp_off [Default Off] 2499 By default, super page will be supported if Intel IOMMU 2500 has the capability. With this option, super page will 2501 not be supported. 2502 sm_on 2503 Enable the Intel IOMMU scalable mode if the hardware 2504 advertises that it has support for the scalable mode 2505 translation. 2506 sm_off 2507 Disallow use of the Intel IOMMU scalable mode. 2508 tboot_noforce [Default Off] 2509 Do not force the Intel IOMMU enabled under tboot. 2510 By default, tboot will force Intel IOMMU on, which 2511 could harm performance of some high-throughput 2512 devices like 40GBit network cards, even if identity 2513 mapping is enabled. 2514 Note that using this option lowers the security 2515 provided by tboot because it makes the system 2516 vulnerable to DMA attacks. 2517 2518 intel_idle.max_cstate= [KNL,HW,ACPI,X86] 2519 0 disables intel_idle and fall back on acpi_idle. 2520 1 to 9 specify maximum depth of C-state. 2521 2522 intel_pstate= [X86,EARLY] 2523 disable 2524 Do not enable intel_pstate as the default 2525 scaling driver for the supported processors 2526 active 2527 Use intel_pstate driver to bypass the scaling 2528 governors layer of cpufreq and provides it own 2529 algorithms for p-state selection. There are two 2530 P-state selection algorithms provided by 2531 intel_pstate in the active mode: powersave and 2532 performance. The way they both operate depends 2533 on whether or not the hardware managed P-states 2534 (HWP) feature has been enabled in the processor 2535 and possibly on the processor model. 2536 passive 2537 Use intel_pstate as a scaling driver, but configure it 2538 to work with generic cpufreq governors (instead of 2539 enabling its internal governor). This mode cannot be 2540 used along with the hardware-managed P-states (HWP) 2541 feature. 2542 force 2543 Enable intel_pstate on systems that prohibit it by default 2544 in favor of acpi-cpufreq. Forcing the intel_pstate driver 2545 instead of acpi-cpufreq may disable platform features, such 2546 as thermal controls and power capping, that rely on ACPI 2547 P-States information being indicated to OSPM and therefore 2548 should be used with caution. This option does not work with 2549 processors that aren't supported by the intel_pstate driver 2550 or on platforms that use pcc-cpufreq instead of acpi-cpufreq. 2551 no_hwp 2552 Do not enable hardware P state control (HWP) 2553 if available. 2554 hwp_only 2555 Only load intel_pstate on systems which support 2556 hardware P state control (HWP) if available. 2557 support_acpi_ppc 2558 Enforce ACPI _PPC performance limits. If the Fixed ACPI 2559 Description Table, specifies preferred power management 2560 profile as "Enterprise Server" or "Performance Server", 2561 then this feature is turned on by default. 2562 per_cpu_perf_limits 2563 Allow per-logical-CPU P-State performance control limits using 2564 cpufreq sysfs interface 2565 no_cas 2566 Do not enable capacity-aware scheduling (CAS) on 2567 hybrid systems 2568 2569 intremap= [X86-64,Intel-IOMMU,EARLY] 2570 on enable Interrupt Remapping (default) 2571 off disable Interrupt Remapping 2572 nosid disable Source ID checking 2573 no_x2apic_optout 2574 BIOS x2APIC opt-out request will be ignored 2575 nopost disable Interrupt Posting 2576 posted_msi 2577 enable MSIs delivered as posted interrupts 2578 2579 iomem= Disable strict checking of access to MMIO memory 2580 strict regions from userspace. 2581 relaxed 2582 2583 iommu= [X86,EARLY] 2584 2585 off 2586 Don't initialize and use any kind of IOMMU. 2587 2588 force 2589 Force the use of the hardware IOMMU even when 2590 it is not actually needed (e.g. because < 3 GB 2591 memory). 2592 2593 noforce 2594 Don't force hardware IOMMU usage when it is not 2595 needed. (default). 2596 2597 biomerge 2598 panic 2599 nopanic 2600 merge 2601 nomerge 2602 2603 soft 2604 Use software bounce buffering (SWIOTLB) (default for 2605 Intel machines). This can be used to prevent the usage 2606 of an available hardware IOMMU. 2607 2608 [X86] 2609 pt 2610 [X86] 2611 nopt 2612 [PPC/POWERNV] 2613 nobypass 2614 Disable IOMMU bypass, using IOMMU for PCI devices. 2615 2616 [X86] 2617 AMD Gart HW IOMMU-specific options: 2618 2619 <size> 2620 Set the size of the remapping area in bytes. 2621 2622 allowed 2623 Overwrite iommu off workarounds for specific chipsets 2624 2625 fullflush 2626 Flush IOMMU on each allocation (default). 2627 2628 nofullflush 2629 Don't use IOMMU fullflush. 2630 2631 memaper[=<order>] 2632 Allocate an own aperture over RAM with size 2633 32MB<<order. (default: order=1, i.e. 64MB) 2634 2635 merge 2636 Do scatter-gather (SG) merging. Implies "force" 2637 (experimental). 2638 2639 nomerge 2640 Don't do scatter-gather (SG) merging. 2641 2642 noaperture 2643 Ask the IOMMU not to touch the aperture for AGP. 2644 2645 noagp 2646 Don't initialize the AGP driver and use full aperture. 2647 2648 panic 2649 Always panic when IOMMU overflows. 2650 2651 iommu.forcedac= [ARM64,X86,EARLY] Control IOVA allocation for PCI devices. 2652 Format: { "0" | "1" } 2653 0 - Try to allocate a 32-bit DMA address first, before 2654 falling back to the full range if needed. 2655 1 - Allocate directly from the full usable range, 2656 forcing Dual Address Cycle for PCI cards supporting 2657 greater than 32-bit addressing. 2658 2659 iommu.strict= [ARM64,X86,S390,EARLY] Configure TLB invalidation behaviour 2660 Format: { "0" | "1" } 2661 0 - Lazy mode. 2662 Request that DMA unmap operations use deferred 2663 invalidation of hardware TLBs, for increased 2664 throughput at the cost of reduced device isolation. 2665 Will fall back to strict mode if not supported by 2666 the relevant IOMMU driver. 2667 1 - Strict mode. 2668 DMA unmap operations invalidate IOMMU hardware TLBs 2669 synchronously. 2670 unset - Use value of CONFIG_IOMMU_DEFAULT_DMA_{LAZY,STRICT}. 2671 Note: on x86, strict mode specified via one of the 2672 legacy driver-specific options takes precedence. 2673 2674 iommu.passthrough= 2675 [ARM64,X86,EARLY] Configure DMA to bypass the IOMMU by default. 2676 Format: { "0" | "1" } 2677 0 - Use IOMMU translation for DMA. 2678 1 - Bypass the IOMMU for DMA. 2679 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH. 2680 2681 io7= [HW] IO7 for Marvel-based Alpha systems 2682 See comment before marvel_specify_io7 in 2683 arch/alpha/kernel/core_marvel.c. 2684 2685 io_delay= [X86,EARLY] I/O delay method 2686 0x80 2687 Standard port 0x80 based delay 2688 0xed 2689 Alternate port 0xed based delay (needed on some systems) 2690 udelay 2691 Simple two microseconds delay 2692 none 2693 No delay 2694 2695 ip= [IP_PNP] 2696 See Documentation/admin-guide/nfs/nfsroot.rst. 2697 2698 ipcmni_extend [KNL,EARLY] Extend the maximum number of unique System V 2699 IPC identifiers from 32,768 to 16,777,216. 2700 2701 ipe.enforce= [IPE] 2702 Format: <bool> 2703 Determine whether IPE starts in permissive (0) or 2704 enforce (1) mode. The default is enforce. 2705 2706 ipe.success_audit= 2707 [IPE] 2708 Format: <bool> 2709 Start IPE with success auditing enabled, emitting 2710 an audit event when a binary is allowed. The default 2711 is 0. 2712 2713 irqaffinity= [SMP] Set the default irq affinity mask 2714 The argument is a cpu list, as described above. 2715 2716 irqchip.gicv2_force_probe= 2717 [ARM,ARM64,EARLY] 2718 Format: <bool> 2719 Force the kernel to look for the second 4kB page 2720 of a GICv2 controller even if the memory range 2721 exposed by the device tree is too small. 2722 2723 irqchip.gicv3_nolpi= 2724 [ARM,ARM64,EARLY] 2725 Force the kernel to ignore the availability of 2726 LPIs (and by consequence ITSs). Intended for system 2727 that use the kernel as a bootloader, and thus want 2728 to let secondary kernels in charge of setting up 2729 LPIs. 2730 2731 irqchip.gicv3_pseudo_nmi= [ARM64,EARLY] 2732 Enables support for pseudo-NMIs in the kernel. This 2733 requires the kernel to be built with 2734 CONFIG_ARM64_PSEUDO_NMI. 2735 2736 irqchip.riscv_imsic_noipi 2737 [RISC-V,EARLY] 2738 Force the kernel to not use IMSIC software injected MSIs 2739 as IPIs. Intended for system where IMSIC is trap-n-emulated, 2740 and thus want to reduce MMIO traps when triggering IPIs 2741 to multiple harts. 2742 2743 irqfixup [HW] 2744 When an interrupt is not handled search all handlers 2745 for it. Intended to get systems with badly broken 2746 firmware running. 2747 2748 irqhandler.duration_warn_us= [KNL] 2749 Warn if an IRQ handler exceeds the specified duration 2750 threshold in microseconds. Useful for identifying 2751 long-running IRQs in the system. 2752 2753 irqpoll [HW] 2754 When an interrupt is not handled search all handlers 2755 for it. Also check all handlers each timer 2756 interrupt. Intended to get systems with badly broken 2757 firmware running. 2758 2759 isapnp= [ISAPNP] 2760 Format: <RDP>,<reset>,<pci_scan>,<verbosity> 2761 2762 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance. 2763 [Deprecated - use cpusets instead] 2764 Format: [flag-list,]<cpu-list> 2765 2766 Specify one or more CPUs to isolate from disturbances 2767 specified in the flag list (default: domain): 2768 2769 nohz 2770 Disable the tick when a single task runs as well as 2771 disabling other kernel noises like having RCU callbacks 2772 offloaded. This is equivalent to the nohz_full parameter. 2773 2774 A residual 1Hz tick is offloaded to workqueues, which you 2775 need to affine to housekeeping through the global 2776 workqueue's affinity configured via the 2777 /sys/devices/virtual/workqueue/cpumask sysfs file, or 2778 by using the 'domain' flag described below. 2779 2780 NOTE: by default the global workqueue runs on all CPUs, 2781 so to protect individual CPUs the 'cpumask' file has to 2782 be configured manually after bootup. 2783 2784 domain 2785 Isolate from the general SMP balancing and scheduling 2786 algorithms. Note that performing domain isolation this way 2787 is irreversible: it's not possible to bring back a CPU to 2788 the domains once isolated through isolcpus. It's strongly 2789 advised to use cpusets instead to disable scheduler load 2790 balancing through the "cpuset.sched_load_balance" file. 2791 It offers a much more flexible interface where CPUs can 2792 move in and out of an isolated set anytime. 2793 2794 You can move a process onto or off an "isolated" CPU via 2795 the CPU affinity syscalls or cpuset. 2796 <cpu number> begins at 0 and the maximum value is 2797 "number of CPUs in system - 1". 2798 2799 managed_irq 2800 2801 Isolate from being targeted by managed interrupts 2802 which have an interrupt mask containing isolated 2803 CPUs. The affinity of managed interrupts is 2804 handled by the kernel and cannot be changed via 2805 the /proc/irq/* interfaces. 2806 2807 This isolation is best effort and only effective 2808 if the automatically assigned interrupt mask of a 2809 device queue contains isolated and housekeeping 2810 CPUs. If housekeeping CPUs are online then such 2811 interrupts are directed to the housekeeping CPU 2812 so that IO submitted on the housekeeping CPU 2813 cannot disturb the isolated CPU. 2814 2815 If a queue's affinity mask contains only isolated 2816 CPUs then this parameter has no effect on the 2817 interrupt routing decision, though interrupts are 2818 only delivered when tasks running on those 2819 isolated CPUs submit IO. IO submitted on 2820 housekeeping CPUs has no influence on those 2821 queues. 2822 2823 The format of <cpu-list> is described above. 2824 2825 iucv= [HW,NET] 2826 2827 ivrs_ioapic [HW,X86-64] 2828 Provide an override to the IOAPIC-ID<->DEVICE-ID 2829 mapping provided in the IVRS ACPI table. 2830 By default, PCI segment is 0, and can be omitted. 2831 2832 For example, to map IOAPIC-ID decimal 10 to 2833 PCI segment 0x1 and PCI device 00:14.0, 2834 write the parameter as: 2835 ivrs_ioapic=10@0001:00:14.0 2836 2837 Deprecated formats: 2838 * To map IOAPIC-ID decimal 10 to PCI device 00:14.0 2839 write the parameter as: 2840 ivrs_ioapic[10]=00:14.0 2841 * To map IOAPIC-ID decimal 10 to PCI segment 0x1 and 2842 PCI device 00:14.0 write the parameter as: 2843 ivrs_ioapic[10]=0001:00:14.0 2844 2845 ivrs_hpet [HW,X86-64] 2846 Provide an override to the HPET-ID<->DEVICE-ID 2847 mapping provided in the IVRS ACPI table. 2848 By default, PCI segment is 0, and can be omitted. 2849 2850 For example, to map HPET-ID decimal 10 to 2851 PCI segment 0x1 and PCI device 00:14.0, 2852 write the parameter as: 2853 ivrs_hpet=10@0001:00:14.0 2854 2855 Deprecated formats: 2856 * To map HPET-ID decimal 0 to PCI device 00:14.0 2857 write the parameter as: 2858 ivrs_hpet[0]=00:14.0 2859 * To map HPET-ID decimal 10 to PCI segment 0x1 and 2860 PCI device 00:14.0 write the parameter as: 2861 ivrs_ioapic[10]=0001:00:14.0 2862 2863 ivrs_acpihid [HW,X86-64] 2864 Provide an override to the ACPI-HID:UID<->DEVICE-ID 2865 mapping provided in the IVRS ACPI table. 2866 By default, PCI segment is 0, and can be omitted. 2867 2868 For example, to map UART-HID:UID AMD0020:0 to 2869 PCI segment 0x1 and PCI device ID 00:14.5, 2870 write the parameter as: 2871 ivrs_acpihid=AMD0020:0@0001:00:14.5 2872 2873 Deprecated formats: 2874 * To map UART-HID:UID AMD0020:0 to PCI segment is 0, 2875 PCI device ID 00:14.5, write the parameter as: 2876 ivrs_acpihid[00:14.5]=AMD0020:0 2877 * To map UART-HID:UID AMD0020:0 to PCI segment 0x1 and 2878 PCI device ID 00:14.5, write the parameter as: 2879 ivrs_acpihid[0001:00:14.5]=AMD0020:0 2880 2881 js= [HW,JOY] Analog joystick 2882 See Documentation/input/joydev/joystick.rst. 2883 2884 kasan_multi_shot 2885 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print 2886 report on every invalid memory access. Without this 2887 parameter KASAN will print report only for the first 2888 invalid access. 2889 2890 keep_bootcon [KNL,EARLY] 2891 Do not unregister boot console at start. This is only 2892 useful for debugging when something happens in the window 2893 between unregistering the boot console and initializing 2894 the real console. 2895 2896 keepinitrd [HW,ARM] See retain_initrd. 2897 2898 kernelcore= [KNL,X86,PPC,EARLY] 2899 Format: nn[KMGTPE] | nn% | "mirror" 2900 This parameter specifies the amount of memory usable by 2901 the kernel for non-movable allocations. The requested 2902 amount is spread evenly throughout all nodes in the 2903 system as ZONE_NORMAL. The remaining memory is used for 2904 movable memory in its own zone, ZONE_MOVABLE. In the 2905 event, a node is too small to have both ZONE_NORMAL and 2906 ZONE_MOVABLE, kernelcore memory will take priority and 2907 other nodes will have a larger ZONE_MOVABLE. 2908 2909 ZONE_MOVABLE is used for the allocation of pages that 2910 may be reclaimed or moved by the page migration 2911 subsystem. Note that allocations like PTEs-from-HighMem 2912 still use the HighMem zone if it exists, and the Normal 2913 zone if it does not. 2914 2915 It is possible to specify the exact amount of memory in 2916 the form of "nn[KMGTPE]", a percentage of total system 2917 memory in the form of "nn%", or "mirror". If "mirror" 2918 option is specified, mirrored (reliable) memory is used 2919 for non-movable allocations and remaining memory is used 2920 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror" 2921 are exclusive, so you cannot specify multiple forms. 2922 2923 kgdbdbgp= [KGDB,HW,EARLY] kgdb over EHCI usb debug port. 2924 Format: <Controller#>[,poll interval] 2925 The controller # is the number of the ehci usb debug 2926 port as it is probed via PCI. The poll interval is 2927 optional and is the number seconds in between 2928 each poll cycle to the debug port in case you need 2929 the functionality for interrupting the kernel with 2930 gdb or control-c on the dbgp connection. When 2931 not using this parameter you use sysrq-g to break into 2932 the kernel debugger. 2933 2934 kgdboc= [KGDB,HW] kgdb over consoles. 2935 Requires a tty driver that supports console polling, 2936 or a supported polling keyboard driver (non-usb). 2937 Serial only format: <serial_device>[,baud] 2938 keyboard only format: kbd 2939 keyboard and serial format: kbd,<serial_device>[,baud] 2940 Optional Kernel mode setting: 2941 kms, kbd format: kms,kbd 2942 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud] 2943 2944 kgdboc_earlycon= [KGDB,HW,EARLY] 2945 If the boot console provides the ability to read 2946 characters and can work in polling mode, you can use 2947 this parameter to tell kgdb to use it as a backend 2948 until the normal console is registered. Intended to 2949 be used together with the kgdboc parameter which 2950 specifies the normal console to transition to. 2951 2952 The name of the early console should be specified 2953 as the value of this parameter. Note that the name of 2954 the early console might be different than the tty 2955 name passed to kgdboc. It's OK to leave the value 2956 blank and the first boot console that implements 2957 read() will be picked. 2958 2959 kgdbwait [KGDB,EARLY] Stop kernel execution and enter the 2960 kernel debugger at the earliest opportunity. 2961 2962 kho= [KEXEC,EARLY] 2963 Format: { "0" | "1" | "off" | "on" | "y" | "n" } 2964 Enables or disables Kexec HandOver. 2965 "0" | "off" | "n" - kexec handover is disabled 2966 "1" | "on" | "y" - kexec handover is enabled 2967 2968 kho_scratch= [KEXEC,EARLY] 2969 Format: ll[KMG],mm[KMG],nn[KMG] | nn% 2970 Defines the size of the KHO scratch region. The KHO 2971 scratch regions are physically contiguous memory 2972 ranges that can only be used for non-kernel 2973 allocations. That way, even when memory is heavily 2974 fragmented with handed over memory, the kexeced 2975 kernel will always have enough contiguous ranges to 2976 bootstrap itself. 2977 2978 It is possible to specify the exact amount of 2979 memory in the form of "ll[KMG],mm[KMG],nn[KMG]" 2980 where the first parameter defines the size of a low 2981 memory scratch area, the second parameter defines 2982 the size of a global scratch area and the third 2983 parameter defines the size of additional per-node 2984 scratch areas. The form "nn%" defines scale factor 2985 (in percents) of memory that was used during boot. 2986 2987 kmac= [MIPS] Korina ethernet MAC address. 2988 Configure the RouterBoard 532 series on-chip 2989 Ethernet adapter MAC address. 2990 2991 kmemleak= [KNL,EARLY] Boot-time kmemleak enable/disable 2992 Valid arguments: on, off 2993 Default: on 2994 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y, 2995 the default is off. 2996 2997 kprobe_event=[probe-list] 2998 [FTRACE] Add kprobe events and enable at boot time. 2999 The probe-list is a semicolon delimited list of probe 3000 definitions. Each definition is same as kprobe_events 3001 interface, but the parameters are comma delimited. 3002 For example, to add a kprobe event on vfs_read with 3003 arg1 and arg2, add to the command line; 3004 3005 kprobe_event=p,vfs_read,$arg1,$arg2 3006 3007 See also Documentation/trace/kprobetrace.rst "Kernel 3008 Boot Parameter" section. 3009 3010 kpti= [ARM64,EARLY] Control page table isolation of 3011 user and kernel address spaces. 3012 Default: enabled on cores which need mitigation. 3013 0: force disabled 3014 1: force enabled 3015 3016 kunit.enable= [KUNIT] Enable executing KUnit tests. Requires 3017 CONFIG_KUNIT to be set to be fully enabled. The 3018 default value can be overridden via 3019 KUNIT_DEFAULT_ENABLED. 3020 Default is 1 (enabled) 3021 3022 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs. 3023 Default is 0 (don't ignore, but inject #GP) 3024 3025 kvm.eager_page_split= 3026 [KVM,X86] Controls whether or not KVM will try to 3027 proactively split all huge pages during dirty logging. 3028 Eager page splitting reduces interruptions to vCPU 3029 execution by eliminating the write-protection faults 3030 and MMU lock contention that would otherwise be 3031 required to split huge pages lazily. 3032 3033 VM workloads that rarely perform writes or that write 3034 only to a small region of VM memory may benefit from 3035 disabling eager page splitting to allow huge pages to 3036 still be used for reads. 3037 3038 The behavior of eager page splitting depends on whether 3039 KVM_DIRTY_LOG_INITIALLY_SET is enabled or disabled. If 3040 disabled, all huge pages in a memslot will be eagerly 3041 split when dirty logging is enabled on that memslot. If 3042 enabled, eager page splitting will be performed during 3043 the KVM_CLEAR_DIRTY ioctl, and only for the pages being 3044 cleared. 3045 3046 Eager page splitting is only supported when kvm.tdp_mmu=Y. 3047 3048 Default is Y (on). 3049 3050 kvm.enable_virt_at_load=[KVM,ARM64,LOONGARCH,MIPS,RISCV,X86] 3051 If enabled, KVM will enable virtualization in hardware 3052 when KVM is loaded, and disable virtualization when KVM 3053 is unloaded (if KVM is built as a module). 3054 3055 If disabled, KVM will dynamically enable and disable 3056 virtualization on-demand when creating and destroying 3057 VMs, i.e. on the 0=>1 and 1=>0 transitions of the 3058 number of VMs. 3059 3060 Enabling virtualization at module load avoids potential 3061 latency for creation of the 0=>1 VM, as KVM serializes 3062 virtualization enabling across all online CPUs. The 3063 "cost" of enabling virtualization when KVM is loaded, 3064 is that doing so may interfere with using out-of-tree 3065 hypervisors that want to "own" virtualization hardware. 3066 3067 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface. 3068 Default is false (don't support). 3069 3070 kvm.nx_huge_pages= 3071 [KVM] Controls the software workaround for the 3072 X86_BUG_ITLB_MULTIHIT bug. 3073 force : Always deploy workaround. 3074 off : Never deploy workaround. 3075 auto : Deploy workaround based on the presence of 3076 X86_BUG_ITLB_MULTIHIT. 3077 3078 Default is 'auto'. 3079 3080 If the software workaround is enabled for the host, 3081 guests do need not to enable it for nested guests. 3082 3083 kvm.nx_huge_pages_recovery_ratio= 3084 [KVM] Controls how many 4KiB pages are periodically zapped 3085 back to huge pages. 0 disables the recovery, otherwise if 3086 the value is N KVM will zap 1/Nth of the 4KiB pages every 3087 period (see below). The default is 60. 3088 3089 kvm.nx_huge_pages_recovery_period_ms= 3090 [KVM] Controls the time period at which KVM zaps 4KiB pages 3091 back to huge pages. If the value is a non-zero N, KVM will 3092 zap a portion (see ratio above) of the pages every N msecs. 3093 If the value is 0 (the default), KVM will pick a period based 3094 on the ratio, such that a page is zapped after 1 hour on average. 3095 3096 kvm-amd.nested= [KVM,AMD] Control nested virtualization feature in 3097 KVM/SVM. Default is 1 (enabled). 3098 3099 kvm-amd.npt= [KVM,AMD] Control KVM's use of Nested Page Tables, 3100 a.k.a. Two-Dimensional Page Tables. Default is 1 3101 (enabled). Disable by KVM if hardware lacks support 3102 for NPT. 3103 3104 kvm-amd.ciphertext_hiding_asids= 3105 [KVM,AMD] Ciphertext hiding prevents disallowed accesses 3106 to SNP private memory from reading ciphertext. Instead, 3107 reads will see constant default values (0xff). 3108 3109 If ciphertext hiding is enabled, the joint SEV-ES and 3110 SEV-SNP ASID space is partitioned into separate SEV-ES 3111 and SEV-SNP ASID ranges, with the SEV-SNP range being 3112 [1..max_snp_asid] and the SEV-ES range being 3113 (max_snp_asid..min_sev_asid), where min_sev_asid is 3114 enumerated by CPUID.0x.8000_001F[EDX]. 3115 3116 A non-zero value enables SEV-SNP ciphertext hiding and 3117 adjusts the ASID ranges for SEV-ES and SEV-SNP guests. 3118 KVM caps the number of SEV-SNP ASIDs at the maximum 3119 possible value, e.g. specifying -1u will assign all 3120 joint SEV-ES and SEV-SNP ASIDs to SEV-SNP. Note, 3121 assigning all joint ASIDs to SEV-SNP, i.e. configuring 3122 max_snp_asid == min_sev_asid-1, will effectively make 3123 SEV-ES unusable. 3124 3125 kvm-arm.mode= 3126 [KVM,ARM,EARLY] Select one of KVM/arm64's modes of 3127 operation. 3128 3129 none: Forcefully disable KVM. 3130 3131 nvhe: Standard nVHE-based mode, without support for 3132 protected guests. 3133 3134 protected: Mode with support for guests whose state is 3135 kept private from the host, using VHE or 3136 nVHE depending on HW support. 3137 3138 nested: VHE-based mode with support for nested 3139 virtualization. Requires at least ARMv8.4 3140 hardware (with FEAT_NV2). 3141 3142 Defaults to VHE/nVHE based on hardware support. Setting 3143 mode to "protected" will disable kexec and hibernation 3144 for the host. To force nVHE on VHE hardware, add 3145 "arm64_sw.hvhe=0 id_aa64mmfr1.vh=0" to the 3146 command-line. 3147 "nested" is experimental and should be used with 3148 extreme caution. 3149 3150 kvm-arm.vgic_v3_group0_trap= 3151 [KVM,ARM,EARLY] Trap guest accesses to GICv3 group-0 3152 system registers 3153 3154 kvm-arm.vgic_v3_group1_trap= 3155 [KVM,ARM,EARLY] Trap guest accesses to GICv3 group-1 3156 system registers 3157 3158 kvm-arm.vgic_v3_common_trap= 3159 [KVM,ARM,EARLY] Trap guest accesses to GICv3 common 3160 system registers 3161 3162 kvm-arm.vgic_v4_enable= 3163 [KVM,ARM,EARLY] Allow use of GICv4 for direct 3164 injection of LPIs. 3165 3166 kvm-arm.wfe_trap_policy= 3167 [KVM,ARM] Control when to set WFE instruction trap for 3168 KVM VMs. Traps are allowed but not guaranteed by the 3169 CPU architecture. 3170 3171 trap: set WFE instruction trap 3172 3173 notrap: clear WFE instruction trap 3174 3175 kvm-arm.wfi_trap_policy= 3176 [KVM,ARM] Control when to set WFI instruction trap for 3177 KVM VMs. Traps are allowed but not guaranteed by the 3178 CPU architecture. 3179 3180 trap: set WFI instruction trap 3181 3182 notrap: clear WFI instruction trap 3183 3184 kvm_cma_resv_ratio=n [PPC,EARLY] 3185 Reserves given percentage from system memory area for 3186 contiguous memory allocation for KVM hash pagetable 3187 allocation. 3188 By default it reserves 5% of total system memory. 3189 Format: <integer> 3190 Default: 5 3191 3192 kvm-intel.ept= [KVM,Intel] Control KVM's use of Extended Page Tables, 3193 a.k.a. Two-Dimensional Page Tables. Default is 1 3194 (enabled). Disable by KVM if hardware lacks support 3195 for EPT. 3196 3197 kvm-intel.emulate_invalid_guest_state= 3198 [KVM,Intel] Control whether to emulate invalid guest 3199 state. Ignored if kvm-intel.enable_unrestricted_guest=1, 3200 as guest state is never invalid for unrestricted 3201 guests. This param doesn't apply to nested guests (L2), 3202 as KVM never emulates invalid L2 guest state. 3203 Default is 1 (enabled). 3204 3205 kvm-intel.flexpriority= 3206 [KVM,Intel] Control KVM's use of FlexPriority feature 3207 (TPR shadow). Default is 1 (enabled). Disable by KVM if 3208 hardware lacks support for it. 3209 3210 kvm-intel.nested= 3211 [KVM,Intel] Control nested virtualization feature in 3212 KVM/VMX. Default is 1 (enabled). 3213 3214 kvm-intel.unrestricted_guest= 3215 [KVM,Intel] Control KVM's use of unrestricted guest 3216 feature (virtualized real and unpaged mode). Default 3217 is 1 (enabled). Disable by KVM if EPT is disabled or 3218 hardware lacks support for it. 3219 3220 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault 3221 CVE-2018-3620. 3222 3223 Valid arguments: never, cond, always 3224 3225 always: L1D cache flush on every VMENTER. 3226 cond: Flush L1D on VMENTER only when the code between 3227 VMEXIT and VMENTER can leak host memory. 3228 never: Disables the mitigation 3229 3230 Default is cond (do L1 cache flush in specific instances) 3231 3232 kvm-intel.vpid= [KVM,Intel] Control KVM's use of Virtual Processor 3233 Identification feature (tagged TLBs). Default is 1 3234 (enabled). Disable by KVM if hardware lacks support 3235 for it. 3236 3237 l1d_flush= [X86,INTEL,EARLY] 3238 Control mitigation for L1D based snooping vulnerability. 3239 3240 Certain CPUs are vulnerable to an exploit against CPU 3241 internal buffers which can forward information to a 3242 disclosure gadget under certain conditions. 3243 3244 In vulnerable processors, the speculatively 3245 forwarded data can be used in a cache side channel 3246 attack, to access data to which the attacker does 3247 not have direct access. 3248 3249 This parameter controls the mitigation. The 3250 options are: 3251 3252 on - enable the interface for the mitigation 3253 3254 l1tf= [X86,EARLY] Control mitigation of the L1TF vulnerability on 3255 affected CPUs 3256 3257 The kernel PTE inversion protection is unconditionally 3258 enabled and cannot be disabled. 3259 3260 full 3261 Provides all available mitigations for the 3262 L1TF vulnerability. Disables SMT and 3263 enables all mitigations in the 3264 hypervisors, i.e. unconditional L1D flush. 3265 3266 SMT control and L1D flush control via the 3267 sysfs interface is still possible after 3268 boot. Hypervisors will issue a warning 3269 when the first VM is started in a 3270 potentially insecure configuration, 3271 i.e. SMT enabled or L1D flush disabled. 3272 3273 full,force 3274 Same as 'full', but disables SMT and L1D 3275 flush runtime control. Implies the 3276 'nosmt=force' command line option. 3277 (i.e. sysfs control of SMT is disabled.) 3278 3279 flush 3280 Leaves SMT enabled and enables the default 3281 hypervisor mitigation, i.e. conditional 3282 L1D flush. 3283 3284 SMT control and L1D flush control via the 3285 sysfs interface is still possible after 3286 boot. Hypervisors will issue a warning 3287 when the first VM is started in a 3288 potentially insecure configuration, 3289 i.e. SMT enabled or L1D flush disabled. 3290 3291 flush,nosmt 3292 3293 Disables SMT and enables the default 3294 hypervisor mitigation. 3295 3296 SMT control and L1D flush control via the 3297 sysfs interface is still possible after 3298 boot. Hypervisors will issue a warning 3299 when the first VM is started in a 3300 potentially insecure configuration, 3301 i.e. SMT enabled or L1D flush disabled. 3302 3303 flush,nowarn 3304 Same as 'flush', but hypervisors will not 3305 warn when a VM is started in a potentially 3306 insecure configuration. 3307 3308 off 3309 Disables hypervisor mitigations and doesn't 3310 emit any warnings. 3311 It also drops the swap size and available 3312 RAM limit restriction on both hypervisor and 3313 bare metal. 3314 3315 Default is 'flush'. 3316 3317 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst 3318 3319 l2cr= [PPC] 3320 3321 l3cr= [PPC] 3322 3323 lapic [X86-32,APIC,EARLY] Enable the local APIC even if BIOS 3324 disabled it. 3325 3326 lapic= [X86,APIC] Do not use TSC deadline 3327 value for LAPIC timer one-shot implementation. Default 3328 back to the programmable timer unit in the LAPIC. 3329 Format: notscdeadline 3330 3331 lapic_timer_c2_ok [X86,APIC,EARLY] trust the local apic timer 3332 in C2 power state. 3333 3334 libata.dma= [LIBATA] DMA control 3335 libata.dma=0 Disable all PATA and SATA DMA 3336 libata.dma=1 PATA and SATA Disk DMA only 3337 libata.dma=2 ATAPI (CDROM) DMA only 3338 libata.dma=4 Compact Flash DMA only 3339 Combinations also work, so libata.dma=3 enables DMA 3340 for disks and CDROMs, but not CFs. 3341 3342 libata.ignore_hpa= [LIBATA] Ignore HPA limit 3343 libata.ignore_hpa=0 keep BIOS limits (default) 3344 libata.ignore_hpa=1 ignore limits, using full disk 3345 3346 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume 3347 when set. 3348 Format: <int> 3349 3350 libata.force= [LIBATA] Force configurations. The format is a comma- 3351 separated list of "[ID:]VAL" where ID is PORT[.DEVICE]. 3352 PORT and DEVICE are decimal numbers matching port, link 3353 or device. Basically, it matches the ATA ID string 3354 printed on console by libata. If the whole ID part is 3355 omitted, the last PORT and DEVICE values are used. If 3356 ID hasn't been specified yet, the configuration applies 3357 to all ports, links and devices. 3358 3359 If only DEVICE is omitted, the parameter applies to 3360 the port and all links and devices behind it. DEVICE 3361 number of 0 either selects the first device or the 3362 first fan-out link behind PMP device. It does not 3363 select the host link. DEVICE number of 15 selects the 3364 host link and device attached to it. 3365 3366 The VAL specifies the configuration to force. As long 3367 as there is no ambiguity, shortcut notation is allowed. 3368 For example, both 1.5 and 1.5G would work for 1.5Gbps. 3369 The following configurations can be forced. 3370 3371 * Cable type: 40c, 80c, short40c, unk, ign or sata. 3372 Any ID with matching PORT is used. 3373 3374 * SATA link speed limit: 1.5Gbps or 3.0Gbps. 3375 3376 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7]. 3377 udma[/][16,25,33,44,66,100,133] notation is also 3378 allowed. 3379 3380 * nohrst, nosrst, norst: suppress hard, soft and both 3381 resets. 3382 3383 * rstonce: only attempt one reset during hot-unplug 3384 link recovery. 3385 3386 * [no]dbdelay: Enable or disable the extra 200ms delay 3387 before debouncing a link PHY and device presence 3388 detection. 3389 3390 * [no]ncq: Turn on or off NCQ. 3391 3392 * [no]ncqtrim: Enable or disable queued DSM TRIM. 3393 3394 * [no]ncqati: Enable or disable NCQ trim on ATI chipset. 3395 3396 * [no]trim: Enable or disable (unqueued) TRIM. 3397 3398 * trim_zero: Indicate that TRIM command zeroes data. 3399 3400 * max_trim_128m: Set 128M maximum trim size limit. 3401 3402 * [no]dma: Turn on or off DMA transfers. 3403 3404 * atapi_dmadir: Enable ATAPI DMADIR bridge support. 3405 3406 * atapi_mod16_dma: Enable the use of ATAPI DMA for 3407 commands that are not a multiple of 16 bytes. 3408 3409 * [no]dmalog: Enable or disable the use of the 3410 READ LOG DMA EXT command to access logs. 3411 3412 * [no]iddevlog: Enable or disable access to the 3413 identify device data log. 3414 3415 * [no]logdir: Enable or disable access to the general 3416 purpose log directory. 3417 3418 * max_sec_128: Set transfer size limit to 128 sectors. 3419 3420 * max_sec_1024: Set or clear transfer size limit to 3421 1024 sectors. 3422 3423 * max_sec_lba48: Set or clear transfer size limit to 3424 65535 sectors. 3425 3426 * external: Mark port as external (hotplug-capable). 3427 3428 * [no]lpm: Enable or disable link power management. 3429 3430 * [no]setxfer: Indicate if transfer speed mode setting 3431 should be skipped. 3432 3433 * [no]fua: Disable or enable FUA (Force Unit Access) 3434 support for devices supporting this feature. 3435 3436 * dump_id: Dump IDENTIFY data. 3437 3438 * disable: Disable this device. 3439 3440 If there are multiple matching configurations changing 3441 the same attribute, the last one is used. 3442 3443 load_ramdisk= [RAM] [Deprecated] 3444 3445 lockd.nlm_grace_period=P [NFS] Assign grace period. 3446 Format: <integer> 3447 3448 lockd.nlm_tcpport=N [NFS] Assign TCP port. 3449 Format: <integer> 3450 3451 lockd.nlm_timeout=T [NFS] Assign timeout value. 3452 Format: <integer> 3453 3454 lockd.nlm_udpport=M [NFS] Assign UDP port. 3455 Format: <integer> 3456 3457 lockdown= [SECURITY,EARLY] 3458 { integrity | confidentiality } 3459 Enable the kernel lockdown feature. If set to 3460 integrity, kernel features that allow userland to 3461 modify the running kernel are disabled. If set to 3462 confidentiality, kernel features that allow userland 3463 to extract confidential information from the kernel 3464 are also disabled. 3465 3466 locktorture.acq_writer_lim= [KNL] 3467 Set the time limit in jiffies for a lock 3468 acquisition. Acquisitions exceeding this limit 3469 will result in a splat once they do complete. 3470 3471 locktorture.bind_readers= [KNL] 3472 Specify the list of CPUs to which the readers are 3473 to be bound. 3474 3475 locktorture.bind_writers= [KNL] 3476 Specify the list of CPUs to which the writers are 3477 to be bound. 3478 3479 locktorture.call_rcu_chains= [KNL] 3480 Specify the number of self-propagating call_rcu() 3481 chains to set up. These are used to ensure that 3482 there is a high probability of an RCU grace period 3483 in progress at any given time. Defaults to 0, 3484 which disables these call_rcu() chains. 3485 3486 locktorture.long_hold= [KNL] 3487 Specify the duration in milliseconds for the 3488 occasional long-duration lock hold time. Defaults 3489 to 100 milliseconds. Select 0 to disable. 3490 3491 locktorture.nested_locks= [KNL] 3492 Specify the maximum lock nesting depth that 3493 locktorture is to exercise, up to a limit of 8 3494 (MAX_NESTED_LOCKS). Specify zero to disable. 3495 Note that this parameter is ineffective on types 3496 of locks that do not support nested acquisition. 3497 3498 locktorture.nreaders_stress= [KNL] 3499 Set the number of locking read-acquisition kthreads. 3500 Defaults to being automatically set based on the 3501 number of online CPUs. 3502 3503 locktorture.nwriters_stress= [KNL] 3504 Set the number of locking write-acquisition kthreads. 3505 3506 locktorture.onoff_holdoff= [KNL] 3507 Set time (s) after boot for CPU-hotplug testing. 3508 3509 locktorture.onoff_interval= [KNL] 3510 Set time (s) between CPU-hotplug operations, or 3511 zero to disable CPU-hotplug testing. 3512 3513 locktorture.rt_boost= [KNL] 3514 Do periodic testing of real-time lock priority 3515 boosting. Select 0 to disable, 1 to boost 3516 only rt_mutex, and 2 to boost unconditionally. 3517 Defaults to 2, which might seem to be an 3518 odd choice, but which should be harmless for 3519 non-real-time spinlocks, due to their disabling 3520 of preemption. Note that non-realtime mutexes 3521 disable boosting. 3522 3523 locktorture.rt_boost_factor= [KNL] 3524 Number that determines how often and for how 3525 long priority boosting is exercised. This is 3526 scaled down by the number of writers, so that the 3527 number of boosts per unit time remains roughly 3528 constant as the number of writers increases. 3529 On the other hand, the duration of each boost 3530 increases with the number of writers. 3531 3532 locktorture.shuffle_interval= [KNL] 3533 Set task-shuffle interval (jiffies). Shuffling 3534 tasks allows some CPUs to go into dyntick-idle 3535 mode during the locktorture test. 3536 3537 locktorture.shutdown_secs= [KNL] 3538 Set time (s) after boot system shutdown. This 3539 is useful for hands-off automated testing. 3540 3541 locktorture.stat_interval= [KNL] 3542 Time (s) between statistics printk()s. 3543 3544 locktorture.stutter= [KNL] 3545 Time (s) to stutter testing, for example, 3546 specifying five seconds causes the test to run for 3547 five seconds, wait for five seconds, and so on. 3548 This tests the locking primitive's ability to 3549 transition abruptly to and from idle. 3550 3551 locktorture.torture_type= [KNL] 3552 Specify the locking implementation to test. 3553 3554 locktorture.verbose= [KNL] 3555 Enable additional printk() statements. 3556 3557 locktorture.writer_fifo= [KNL] 3558 Run the write-side locktorture kthreads at 3559 sched_set_fifo() real-time priority. 3560 3561 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver 3562 Format: <irq> 3563 3564 loglevel= [KNL,EARLY] 3565 All Kernel Messages with a loglevel smaller than the 3566 console loglevel will be printed to the console. It can 3567 also be changed with klogd or other programs. The 3568 loglevels are defined as follows: 3569 3570 0 (KERN_EMERG) system is unusable 3571 1 (KERN_ALERT) action must be taken immediately 3572 2 (KERN_CRIT) critical conditions 3573 3 (KERN_ERR) error conditions 3574 4 (KERN_WARNING) warning conditions 3575 5 (KERN_NOTICE) normal but significant condition 3576 6 (KERN_INFO) informational 3577 7 (KERN_DEBUG) debug-level messages 3578 3579 log_buf_len=n[KMG] [KNL,EARLY] 3580 Sets the size of the printk ring buffer, in bytes. 3581 n must be a power of two and greater than the 3582 minimal size. The minimal size is defined by 3583 LOG_BUF_SHIFT kernel config parameter. There 3584 is also CONFIG_LOG_CPU_MAX_BUF_SHIFT config 3585 parameter that allows to increase the default size 3586 depending on the number of CPUs. See init/Kconfig 3587 for more details. 3588 3589 logo.nologo [FB] Disables display of the built-in Linux logo. 3590 This may be used to provide more screen space for 3591 kernel log messages and is useful when debugging 3592 kernel boot problems. 3593 3594 lp=0 [LP] Specify parallel ports to use, e.g, 3595 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses 3596 lp=reset first parallel port). 'lp=0' disables the 3597 lp=auto printer driver. 'lp=reset' (which can be 3598 specified in addition to the ports) causes 3599 attached printers to be reset. Using 3600 lp=port1,port2,... specifies the parallel ports 3601 to associate lp devices with, starting with 3602 lp0. A port specification may be 'none' to skip 3603 that lp device, or a parport name such as 3604 'parport0'. Specifying 'lp=auto' instead of a 3605 port specification list means that device IDs 3606 from each port should be examined, to see if 3607 an IEEE 1284-compliant printer is attached; if 3608 so, the driver will manage that printer. 3609 See also header of drivers/char/lp.c. 3610 3611 lpj=n [KNL] 3612 Sets loops_per_jiffy to given constant, thus avoiding 3613 time-consuming boot-time autodetection (up to 250 ms per 3614 CPU). 0 enables autodetection (default). To determine 3615 the correct value for your kernel, boot with normal 3616 autodetection and see what value is printed. Note that 3617 on SMP systems the preset will be applied to all CPUs, 3618 which is likely to cause problems if your CPUs need 3619 significantly divergent settings. An incorrect value 3620 will cause delays in the kernel to be wrong, leading to 3621 unpredictable I/O errors and other breakage. Although 3622 unlikely, in the extreme case this might damage your 3623 hardware. 3624 3625 lsm.debug [SECURITY] Enable LSM initialization debugging output. 3626 3627 lsm=lsm1,...,lsmN 3628 [SECURITY] Choose order of LSM initialization. This 3629 overrides CONFIG_LSM, and the "security=" parameter. 3630 3631 machtype= [Loongson] Share the same kernel image file between 3632 different yeeloong laptops. 3633 Example: machtype=lemote-yeeloong-2f-7inch 3634 3635 maxcpus= [SMP,EARLY] Maximum number of processors that an SMP kernel 3636 will bring up during bootup. maxcpus=n : n >= 0 limits 3637 the kernel to bring up 'n' processors. Surely after 3638 bootup you can bring up the other plugged cpu by executing 3639 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus 3640 only takes effect during system bootup. 3641 While n=0 is a special case, it is equivalent to "nosmp", 3642 which also disables the IO APIC. 3643 3644 max_loop= [LOOP] The number of loop block devices that get 3645 (loop.max_loop) unconditionally pre-created at init time. The default 3646 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead 3647 of statically allocating a predefined number, loop 3648 devices can be requested on-demand with the 3649 /dev/loop-control interface. 3650 3651 mce= [X86-{32,64}] 3652 3653 Please see Documentation/arch/x86/x86_64/machinecheck.rst for sysfs runtime tunables. 3654 3655 off 3656 disable machine check 3657 3658 no_cmci 3659 disable CMCI(Corrected Machine Check Interrupt) that 3660 Intel processor supports. Usually this disablement is 3661 not recommended, but it might be handy if your 3662 hardware is misbehaving. 3663 3664 Note that you'll get more problems without CMCI than 3665 with due to the shared banks, i.e. you might get 3666 duplicated error logs. 3667 3668 dont_log_ce 3669 don't make logs for corrected errors. All events 3670 reported as corrected are silently cleared by OS. This 3671 option will be useful if you have no interest in any 3672 of corrected errors. 3673 3674 ignore_ce 3675 disable features for corrected errors, e.g. 3676 polling timer and CMCI. All events reported as 3677 corrected are not cleared by OS and remained in its 3678 error banks. 3679 3680 Usually this disablement is not recommended, however 3681 if there is an agent checking/clearing corrected 3682 errors (e.g. BIOS or hardware monitoring 3683 applications), conflicting with OS's error handling, 3684 and you cannot deactivate the agent, then this option 3685 will be a help. 3686 3687 no_lmce 3688 do not opt-in to Local MCE delivery. Use legacy method 3689 to broadcast MCEs. 3690 3691 bootlog 3692 enable logging of machine checks left over from 3693 booting. Disabled by default on AMD Fam10h and older 3694 because some BIOS leave bogus ones. 3695 3696 If your BIOS doesn't do that it's a good idea to 3697 enable though to make sure you log even machine check 3698 events that result in a reboot. On Intel systems it is 3699 enabled by default. 3700 3701 nobootlog 3702 disable boot machine check logging. 3703 3704 monarchtimeout (number) 3705 sets the time in us to wait for other CPUs on machine 3706 checks. 0 to disable. 3707 3708 bios_cmci_threshold 3709 don't overwrite the bios-set CMCI threshold. This boot 3710 option prevents Linux from overwriting the CMCI 3711 threshold set by the bios. Without this option, Linux 3712 always sets the CMCI threshold to 1. Enabling this may 3713 make memory predictive failure analysis less effective 3714 if the bios sets thresholds for memory errors since we 3715 will not see details for all errors. 3716 3717 recovery 3718 force-enable recoverable machine check code paths 3719 3720 Everything else is in sysfs now. 3721 3722 3723 md= [HW] RAID subsystems devices and level 3724 See Documentation/admin-guide/md.rst. 3725 3726 mdacon= [MDA] 3727 Format: <first>,<last> 3728 Specifies range of consoles to be captured by the MDA. 3729 3730 mds= [X86,INTEL,EARLY] 3731 Control mitigation for the Micro-architectural Data 3732 Sampling (MDS) vulnerability. 3733 3734 Certain CPUs are vulnerable to an exploit against CPU 3735 internal buffers which can forward information to a 3736 disclosure gadget under certain conditions. 3737 3738 In vulnerable processors, the speculatively 3739 forwarded data can be used in a cache side channel 3740 attack, to access data to which the attacker does 3741 not have direct access. 3742 3743 This parameter controls the MDS mitigation. The 3744 options are: 3745 3746 full - Enable MDS mitigation on vulnerable CPUs 3747 full,nosmt - Enable MDS mitigation and disable 3748 SMT on vulnerable CPUs 3749 off - Unconditionally disable MDS mitigation 3750 3751 On TAA-affected machines, mds=off can be prevented by 3752 an active TAA mitigation as both vulnerabilities are 3753 mitigated with the same mechanism so in order to disable 3754 this mitigation, you need to specify tsx_async_abort=off 3755 too. 3756 3757 Not specifying this option is equivalent to 3758 mds=full. 3759 3760 For details see: Documentation/admin-guide/hw-vuln/mds.rst 3761 3762 mem=nn[KMG] [HEXAGON,EARLY] Set the memory size. 3763 Must be specified, otherwise memory size will be 0. 3764 3765 mem=nn[KMG] [KNL,BOOT,EARLY] Force usage of a specific amount 3766 of memory Amount of memory to be used in cases 3767 as follows: 3768 3769 1 for test; 3770 2 when the kernel is not able to see the whole system memory; 3771 3 memory that lies after 'mem=' boundary is excluded from 3772 the hypervisor, then assigned to KVM guests. 3773 4 to limit the memory available for kdump kernel. 3774 3775 [ARC,MICROBLAZE] - the limit applies only to low memory, 3776 high memory is not affected. 3777 3778 [ARM64] - only limits memory covered by the linear 3779 mapping. The NOMAP regions are not affected. 3780 3781 [X86] Work as limiting max address. Use together 3782 with memmap= to avoid physical address space collisions. 3783 Without memmap= PCI devices could be placed at addresses 3784 belonging to unused RAM. 3785 3786 Note that this only takes effects during boot time since 3787 in above case 3, memory may need be hot added after boot 3788 if system memory of hypervisor is not sufficient. 3789 3790 mem=nn[KMG]@ss[KMG] 3791 [ARM,MIPS,EARLY] - override the memory layout 3792 reported by firmware. 3793 Define a memory region of size nn[KMG] starting at 3794 ss[KMG]. 3795 Multiple different regions can be specified with 3796 multiple mem= parameters on the command line. 3797 3798 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel 3799 memory. 3800 3801 memblock=debug [KNL,EARLY] Enable memblock debug messages. 3802 3803 memchunk=nn[KMG] 3804 [KNL,SH] Allow user to override the default size for 3805 per-device physically contiguous DMA buffers. 3806 3807 memhp_default_state=online/offline/online_kernel/online_movable 3808 [KNL] Set the initial state for the memory hotplug 3809 onlining policy. If not specified, the default value is 3810 set according to the 3811 CONFIG_MHP_DEFAULT_ONLINE_TYPE kernel config 3812 options. 3813 See Documentation/admin-guide/mm/memory-hotplug.rst. 3814 3815 memmap=exactmap [KNL,X86,EARLY] Enable setting of an exact 3816 E820 memory map, as specified by the user. 3817 Such memmap=exactmap lines can be constructed based on 3818 BIOS output or other requirements. See the memmap=nn@ss 3819 option description. 3820 3821 memmap=nn[KMG]@ss[KMG] 3822 [KNL, X86,MIPS,XTENSA,EARLY] Force usage of a specific region of memory. 3823 Region of memory to be used is from ss to ss+nn. 3824 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG], 3825 which limits max address to nn[KMG]. 3826 Multiple different regions can be specified, 3827 comma delimited. 3828 Example: 3829 memmap=100M@2G,100M#3G,1G!1024G 3830 3831 memmap=nn[KMG]#ss[KMG] 3832 [KNL,ACPI,EARLY] Mark specific memory as ACPI data. 3833 Region of memory to be marked is from ss to ss+nn. 3834 3835 memmap=nn[KMG]$ss[KMG] 3836 [KNL,ACPI,EARLY] Mark specific memory as reserved. 3837 Region of memory to be reserved is from ss to ss+nn. 3838 Example: Exclude memory from 0x18690000-0x1869ffff 3839 memmap=64K$0x18690000 3840 or 3841 memmap=0x10000$0x18690000 3842 Some bootloaders may need an escape character before '$', 3843 like Grub2, otherwise '$' and the following number 3844 will be eaten. 3845 3846 memmap=nn[KMG]!ss[KMG,EARLY] 3847 [KNL,X86] Mark specific memory as protected. 3848 Region of memory to be used, from ss to ss+nn. 3849 The memory region may be marked as e820 type 12 (0xc) 3850 and is NVDIMM or ADR memory. 3851 3852 memmap=<size>%<offset>-<oldtype>+<newtype> 3853 [KNL,ACPI,EARLY] Convert memory within the specified region 3854 from <oldtype> to <newtype>. If "-<oldtype>" is left 3855 out, the whole region will be marked as <newtype>, 3856 even if previously unavailable. If "+<newtype>" is left 3857 out, matching memory will be removed. Types are 3858 specified as e820 types, e.g., 1 = RAM, 2 = reserved, 3859 3 = ACPI, 12 = PRAM. 3860 3861 memory_corruption_check=0/1 [X86,EARLY] 3862 Some BIOSes seem to corrupt the first 64k of 3863 memory when doing things like suspend/resume. 3864 Setting this option will scan the memory 3865 looking for corruption. Enabling this will 3866 both detect corruption and prevent the kernel 3867 from using the memory being corrupted. 3868 However, it's intended as a diagnostic tool; if 3869 repeatable BIOS-originated corruption always 3870 affects the same memory, you can use memmap= 3871 to prevent the kernel from using that memory. 3872 3873 memory_corruption_check_size=size [X86,EARLY] 3874 By default it checks for corruption in the low 3875 64k, making this memory unavailable for normal 3876 use. Use this parameter to scan for 3877 corruption in more or less memory. 3878 3879 memory_corruption_check_period=seconds [X86,EARLY] 3880 By default it checks for corruption every 60 3881 seconds. Use this parameter to check at some 3882 other rate. 0 disables periodic checking. 3883 3884 memory_hotplug.memmap_on_memory 3885 [KNL,X86,ARM] Boolean flag to enable this feature. 3886 Format: {on | off (default)} 3887 When enabled, runtime hotplugged memory will 3888 allocate its internal metadata (struct pages, 3889 those vmemmap pages cannot be optimized even 3890 if hugetlb_free_vmemmap is enabled) from the 3891 hotadded memory which will allow to hotadd a 3892 lot of memory without requiring additional 3893 memory to do so. 3894 This feature is disabled by default because it 3895 has some implication on large (e.g. GB) 3896 allocations in some configurations (e.g. small 3897 memory blocks). 3898 The state of the flag can be read in 3899 /sys/module/memory_hotplug/parameters/memmap_on_memory. 3900 Note that even when enabled, there are a few cases where 3901 the feature is not effective. 3902 3903 memtest= [KNL,X86,ARM,M68K,PPC,RISCV,EARLY] Enable memtest 3904 Format: <integer> 3905 default : 0 <disable> 3906 Specifies the number of memtest passes to be 3907 performed. Each pass selects another test 3908 pattern from a given set of patterns. Memtest 3909 fills the memory with this pattern, validates 3910 memory contents and reserves bad memory 3911 regions that are detected. 3912 3913 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control 3914 Valid arguments: on, off 3915 Default: off 3916 mem_encrypt=on: Activate SME 3917 mem_encrypt=off: Do not activate SME 3918 3919 Refer to Documentation/virt/kvm/x86/amd-memory-encryption.rst 3920 for details on when memory encryption can be activated. 3921 3922 mem_sleep_default= [SUSPEND] Default system suspend mode: 3923 s2idle - Suspend-To-Idle 3924 shallow - Power-On Suspend or equivalent (if supported) 3925 deep - Suspend-To-RAM or equivalent (if supported) 3926 See Documentation/admin-guide/pm/sleep-states.rst. 3927 3928 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when 3929 the BIOS has incorrectly applied a workaround. TinyBIOS 3930 version 0.98 is known to be affected, 0.99 fixes the 3931 problem by letting the user disable the workaround. 3932 3933 mga= [HW,DRM] 3934 3935 microcode= [X86] Control the behavior of the microcode loader. 3936 Available options, comma separated: 3937 3938 base_rev=X - with <X> with format: <u32> 3939 Set the base microcode revision of each thread when in 3940 debug mode. 3941 3942 dis_ucode_ldr: disable the microcode loader 3943 3944 force_minrev: 3945 Enable or disable the microcode minimal revision 3946 enforcement for the runtime microcode loader. 3947 3948 mini2440= [ARM,HW,KNL] 3949 Format:[0..2][b][c][t] 3950 Default: "0tb" 3951 MINI2440 configuration specification: 3952 0 - The attached screen is the 3.5" TFT 3953 1 - The attached screen is the 7" TFT 3954 2 - The VGA Shield is attached (1024x768) 3955 Leaving out the screen size parameter will not load 3956 the TFT driver, and the framebuffer will be left 3957 unconfigured. 3958 b - Enable backlight. The TFT backlight pin will be 3959 linked to the kernel VESA blanking code and a GPIO 3960 LED. This parameter is not necessary when using the 3961 VGA shield. 3962 c - Enable the s3c camera interface. 3963 t - Reserved for enabling touchscreen support. The 3964 touchscreen support is not enabled in the mainstream 3965 kernel as of 2.6.30, a preliminary port can be found 3966 in the "bleeding edge" mini2440 support kernel at 3967 https://repo.or.cz/w/linux-2.6/mini2440.git 3968 3969 mitigations= 3970 [X86,PPC,S390,ARM64,EARLY] Control optional mitigations for 3971 CPU vulnerabilities. This is a set of curated, 3972 arch-independent options, each of which is an 3973 aggregation of existing arch-specific options. 3974 3975 Note, "mitigations" is supported if and only if the 3976 kernel was built with CPU_MITIGATIONS=y. 3977 3978 off 3979 Disable all optional CPU mitigations. This 3980 improves system performance, but it may also 3981 expose users to several CPU vulnerabilities. 3982 Equivalent to: if nokaslr then kpti=0 [ARM64] 3983 gather_data_sampling=off [X86] 3984 indirect_target_selection=off [X86] 3985 kvm.nx_huge_pages=off [X86] 3986 l1tf=off [X86] 3987 mds=off [X86] 3988 mmio_stale_data=off [X86] 3989 no_entry_flush [PPC] 3990 no_uaccess_flush [PPC] 3991 nobp=0 [S390] 3992 nopti [X86,PPC] 3993 nospectre_bhb [ARM64] 3994 nospectre_v1 [X86,PPC] 3995 nospectre_v2 [X86,PPC,S390,ARM64] 3996 reg_file_data_sampling=off [X86] 3997 retbleed=off [X86] 3998 spec_rstack_overflow=off [X86] 3999 spec_store_bypass_disable=off [X86,PPC] 4000 spectre_bhi=off [X86] 4001 spectre_v2_user=off [X86] 4002 srbds=off [X86,INTEL] 4003 ssbd=force-off [ARM64] 4004 tsa=off [X86,AMD] 4005 tsx_async_abort=off [X86] 4006 vmscape=off [X86] 4007 4008 Exceptions: 4009 This does not have any effect on 4010 kvm.nx_huge_pages when 4011 kvm.nx_huge_pages=force. 4012 4013 auto (default) 4014 Mitigate all CPU vulnerabilities, but leave SMT 4015 enabled, even if it's vulnerable. This is for 4016 users who don't want to be surprised by SMT 4017 getting disabled across kernel upgrades, or who 4018 have other ways of avoiding SMT-based attacks. 4019 Equivalent to: (default behavior) 4020 4021 auto,nosmt 4022 Mitigate all CPU vulnerabilities, disabling SMT 4023 if needed. This is for users who always want to 4024 be fully mitigated, even if it means losing SMT. 4025 Equivalent to: l1tf=flush,nosmt [X86] 4026 mds=full,nosmt [X86] 4027 tsx_async_abort=full,nosmt [X86] 4028 mmio_stale_data=full,nosmt [X86] 4029 retbleed=auto,nosmt [X86] 4030 4031 [X86] After one of the above options, additionally 4032 supports attack-vector based controls as documented in 4033 Documentation/admin-guide/hw-vuln/attack_vector_controls.rst 4034 4035 mminit_loglevel= 4036 [KNL,EARLY] When CONFIG_DEBUG_MEMORY_INIT is set, this 4037 parameter allows control of the logging verbosity for 4038 the additional memory initialisation checks. A value 4039 of 0 disables mminit logging and a level of 4 will 4040 log everything. Information is printed at KERN_DEBUG 4041 so loglevel=8 may also need to be specified. 4042 4043 mmio_stale_data= 4044 [X86,INTEL,EARLY] Control mitigation for the Processor 4045 MMIO Stale Data vulnerabilities. 4046 4047 Processor MMIO Stale Data is a class of 4048 vulnerabilities that may expose data after an MMIO 4049 operation. Exposed data could originate or end in 4050 the same CPU buffers as affected by MDS and TAA. 4051 Therefore, similar to MDS and TAA, the mitigation 4052 is to clear the affected CPU buffers. 4053 4054 This parameter controls the mitigation. The 4055 options are: 4056 4057 full - Enable mitigation on vulnerable CPUs 4058 4059 full,nosmt - Enable mitigation and disable SMT on 4060 vulnerable CPUs. 4061 4062 off - Unconditionally disable mitigation 4063 4064 On MDS or TAA affected machines, 4065 mmio_stale_data=off can be prevented by an active 4066 MDS or TAA mitigation as these vulnerabilities are 4067 mitigated with the same mechanism so in order to 4068 disable this mitigation, you need to specify 4069 mds=off and tsx_async_abort=off too. 4070 4071 Not specifying this option is equivalent to 4072 mmio_stale_data=full. 4073 4074 For details see: 4075 Documentation/admin-guide/hw-vuln/processor_mmio_stale_data.rst 4076 4077 <module>.async_probe[=<bool>] [KNL] 4078 If no <bool> value is specified or if the value 4079 specified is not a valid <bool>, enable asynchronous 4080 probe on this module. Otherwise, enable/disable 4081 asynchronous probe on this module as indicated by the 4082 <bool> value. See also: module.async_probe 4083 4084 module.async_probe=<bool> 4085 [KNL] When set to true, modules will use async probing 4086 by default. To enable/disable async probing for a 4087 specific module, use the module specific control that 4088 is documented under <module>.async_probe. When both 4089 module.async_probe and <module>.async_probe are 4090 specified, <module>.async_probe takes precedence for 4091 the specific module. 4092 4093 module.enable_dups_trace 4094 [KNL] When CONFIG_MODULE_DEBUG_AUTOLOAD_DUPS is set, 4095 this means that duplicate request_module() calls will 4096 trigger a WARN_ON() instead of a pr_warn(). Note that 4097 if MODULE_DEBUG_AUTOLOAD_DUPS_TRACE is set, WARN_ON()s 4098 will always be issued and this option does nothing. 4099 module.sig_enforce 4100 [KNL] When CONFIG_MODULE_SIG is set, this means that 4101 modules without (valid) signatures will fail to load. 4102 Note that if CONFIG_MODULE_SIG_FORCE is set, that 4103 is always true, so this option does nothing. 4104 4105 module_blacklist= [KNL] Do not load a comma-separated list of 4106 modules. Useful for debugging problem modules. 4107 4108 mousedev.tap_time= 4109 [MOUSE] Maximum time between finger touching and 4110 leaving touchpad surface for touch to be considered 4111 a tap and be reported as a left button click (for 4112 touchpads working in absolute mode only). 4113 Format: <msecs> 4114 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices 4115 reporting absolute coordinates, such as tablets 4116 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices 4117 reporting absolute coordinates, such as tablets 4118 4119 movablecore= [KNL,X86,PPC,EARLY] 4120 Format: nn[KMGTPE] | nn% 4121 This parameter is the complement to kernelcore=, it 4122 specifies the amount of memory used for migratable 4123 allocations. If both kernelcore and movablecore is 4124 specified, then kernelcore will be at *least* the 4125 specified value but may be more. If movablecore on its 4126 own is specified, the administrator must be careful 4127 that the amount of memory usable for all allocations 4128 is not too small. 4129 4130 movable_node [KNL,EARLY] Boot-time switch to make hotplugable memory 4131 NUMA nodes to be movable. This means that the memory 4132 of such nodes will be usable only for movable 4133 allocations which rules out almost all kernel 4134 allocations. Use with caution! 4135 4136 MTD_Partition= [MTD] 4137 Format: <name>,<region-number>,<size>,<offset> 4138 4139 MTD_Region= [MTD] Format: 4140 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>] 4141 4142 mtdparts= [MTD] 4143 See drivers/mtd/parsers/cmdlinepart.c 4144 4145 mtouchusb.raw_coordinates= 4146 [HW] Make the MicroTouch USB driver use raw coordinates 4147 ('y', default) or cooked coordinates ('n') 4148 4149 mtrr=debug [X86,EARLY] 4150 Enable printing debug information related to MTRR 4151 registers at boot time. 4152 4153 mtrr_chunk_size=nn[KMG,X86,EARLY] 4154 used for mtrr cleanup. It is largest continuous chunk 4155 that could hold holes aka. UC entries. 4156 4157 mtrr_gran_size=nn[KMG,X86,EARLY] 4158 Used for mtrr cleanup. It is granularity of mtrr block. 4159 Default is 1. 4160 Large value could prevent small alignment from 4161 using up MTRRs. 4162 4163 mtrr_spare_reg_nr=n [X86,EARLY] 4164 Format: <integer> 4165 Range: 0,7 : spare reg number 4166 Default : 1 4167 Used for mtrr cleanup. It is spare mtrr entries number. 4168 Set to 2 or more if your graphical card needs more. 4169 4170 multitce=off [PPC] This parameter disables the use of the pSeries 4171 firmware feature for updating multiple TCE entries 4172 at a time. 4173 4174 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card 4175 4176 netdev= [NET] Network devices parameters 4177 Format: <irq>,<io>,<mem_start>,<mem_end>,<name> 4178 Note that mem_start is often overloaded to mean 4179 something different and driver-specific. 4180 This usage is only documented in each driver source 4181 file if at all. 4182 4183 netpoll.carrier_timeout= 4184 [NET] Specifies amount of time (in seconds) that 4185 netpoll should wait for a carrier. By default netpoll 4186 waits 4 seconds. 4187 4188 nf_conntrack.acct= 4189 [NETFILTER] Enable connection tracking flow accounting 4190 0 to disable accounting 4191 1 to enable accounting 4192 Default value is 0. 4193 4194 nfs.cache_getent= 4195 [NFS] sets the pathname to the program which is used 4196 to update the NFS client cache entries. 4197 4198 nfs.cache_getent_timeout= 4199 [NFS] sets the timeout after which an attempt to 4200 update a cache entry is deemed to have failed. 4201 4202 nfs.callback_nr_threads= 4203 [NFSv4] set the total number of threads that the 4204 NFS client will assign to service NFSv4 callback 4205 requests. 4206 4207 nfs.callback_tcpport= 4208 [NFS] set the TCP port on which the NFSv4 callback 4209 channel should listen. 4210 4211 nfs.delay_retrans= 4212 [NFS] specifies the number of times the NFSv4 client 4213 retries the request before returning an EAGAIN error, 4214 after a reply of NFS4ERR_DELAY from the server. 4215 Only applies if the softerr mount option is enabled, 4216 and the specified value is >= 0. 4217 4218 nfs.enable_ino64= 4219 [NFS] enable 64-bit inode numbers. 4220 If zero, the NFS client will fake up a 32-bit inode 4221 number for the readdir() and stat() syscalls instead 4222 of returning the full 64-bit number. 4223 The default is to return 64-bit inode numbers. 4224 4225 nfs.idmap_cache_timeout= 4226 [NFS] set the maximum lifetime for idmapper cache 4227 entries. 4228 4229 nfs.max_session_cb_slots= 4230 [NFSv4.1] Sets the maximum number of session 4231 slots the client will assign to the callback 4232 channel. This determines the maximum number of 4233 callbacks the client will process in parallel for 4234 a particular server. 4235 4236 nfs.max_session_slots= 4237 [NFSv4.1] Sets the maximum number of session slots 4238 the client will attempt to negotiate with the server. 4239 This limits the number of simultaneous RPC requests 4240 that the client can send to the NFSv4.1 server. 4241 Note that there is little point in setting this 4242 value higher than the max_tcp_slot_table_limit. 4243 4244 nfs.nfs4_disable_idmapping= 4245 [NFSv4] When set to the default of '1', this option 4246 ensures that both the RPC level authentication 4247 scheme and the NFS level operations agree to use 4248 numeric uids/gids if the mount is using the 4249 'sec=sys' security flavour. In effect it is 4250 disabling idmapping, which can make migration from 4251 legacy NFSv2/v3 systems to NFSv4 easier. 4252 Servers that do not support this mode of operation 4253 will be autodetected by the client, and it will fall 4254 back to using the idmapper. 4255 To turn off this behaviour, set the value to '0'. 4256 4257 nfs.nfs4_unique_id= 4258 [NFS4] Specify an additional fixed unique ident- 4259 ification string that NFSv4 clients can insert into 4260 their nfs_client_id4 string. This is typically a 4261 UUID that is generated at system install time. 4262 4263 nfs.recover_lost_locks= 4264 [NFSv4] Attempt to recover locks that were lost due 4265 to a lease timeout on the server. Please note that 4266 doing this risks data corruption, since there are 4267 no guarantees that the file will remain unchanged 4268 after the locks are lost. 4269 If you want to enable the kernel legacy behaviour of 4270 attempting to recover these locks, then set this 4271 parameter to '1'. 4272 The default parameter value of '0' causes the kernel 4273 not to attempt recovery of lost locks. 4274 4275 nfs.send_implementation_id= 4276 [NFSv4.1] Send client implementation identification 4277 information in exchange_id requests. 4278 If zero, no implementation identification information 4279 will be sent. 4280 The default is to send the implementation identification 4281 information. 4282 4283 nfs4.layoutstats_timer= 4284 [NFSv4.2] Change the rate at which the kernel sends 4285 layoutstats to the pNFS metadata server. 4286 4287 Setting this to value to 0 causes the kernel to use 4288 whatever value is the default set by the layout 4289 driver. A non-zero value sets the minimum interval 4290 in seconds between layoutstats transmissions. 4291 4292 nfsd.inter_copy_offload_enable= 4293 [NFSv4.2] When set to 1, the server will support 4294 server-to-server copies for which this server is 4295 the destination of the copy. 4296 4297 nfsd.nfs4_disable_idmapping= 4298 [NFSv4] When set to the default of '1', the NFSv4 4299 server will return only numeric uids and gids to 4300 clients using auth_sys, and will accept numeric uids 4301 and gids from such clients. This is intended to ease 4302 migration from NFSv2/v3. 4303 4304 nfsd.nfsd4_ssc_umount_timeout= 4305 [NFSv4.2] When used as the destination of a 4306 server-to-server copy, knfsd temporarily mounts 4307 the source server. It caches the mount in case 4308 it will be needed again, and discards it if not 4309 used for the number of milliseconds specified by 4310 this parameter. 4311 4312 nfsaddrs= [NFS] Deprecated. Use ip= instead. 4313 See Documentation/admin-guide/nfs/nfsroot.rst. 4314 4315 nfsroot= [NFS] nfs root filesystem for disk-less boxes. 4316 See Documentation/admin-guide/nfs/nfsroot.rst. 4317 4318 nfsrootdebug [NFS] enable nfsroot debugging messages. 4319 See Documentation/admin-guide/nfs/nfsroot.rst. 4320 4321 nmi_backtrace.backtrace_idle [KNL] 4322 Dump stacks even of idle CPUs in response to an 4323 NMI stack-backtrace request. 4324 4325 nmi_debug= [KNL,SH] Specify one or more actions to take 4326 when a NMI is triggered. 4327 Format: [state][,regs][,debounce][,die] 4328 4329 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels 4330 Format: [panic,][nopanic,][rNNN,][num] 4331 Valid num: 0 or 1 4332 0 - turn hardlockup detector in nmi_watchdog off 4333 1 - turn hardlockup detector in nmi_watchdog on 4334 rNNN - configure the watchdog with raw perf event 0xNNN 4335 4336 When panic is specified, panic when an NMI watchdog 4337 timeout occurs (or 'nopanic' to not panic on an NMI 4338 watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set) 4339 To disable both hard and soft lockup detectors, 4340 please see 'nowatchdog'. 4341 This is useful when you use a panic=... timeout and 4342 need the box quickly up again. 4343 4344 These settings can be accessed at runtime via 4345 the nmi_watchdog and hardlockup_panic sysctls. 4346 4347 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths 4348 emulation library even if a 387 maths coprocessor 4349 is present. 4350 4351 no4lvl [RISCV,EARLY] Disable 4-level and 5-level paging modes. 4352 Forces kernel to use 3-level paging instead. 4353 4354 no5lvl [X86-64,RISCV,EARLY] Disable 5-level paging mode. Forces 4355 kernel to use 4-level paging instead. 4356 4357 noalign [KNL,ARM] 4358 4359 noapic [SMP,APIC,EARLY] Tells the kernel to not make use of any 4360 IOAPICs that may be present in the system. 4361 4362 noapictimer [APIC,X86] Don't set up the APIC timer 4363 4364 noautogroup Disable scheduler automatic task group creation. 4365 4366 nocache [ARM,EARLY] 4367 4368 no_console_suspend 4369 [HW] Never suspend the console 4370 Disable suspending of consoles during suspend and 4371 hibernate operations. Once disabled, debugging 4372 messages can reach various consoles while the rest 4373 of the system is being put to sleep (ie, while 4374 debugging driver suspend/resume hooks). This may 4375 not work reliably with all consoles, but is known 4376 to work with serial and VGA consoles. 4377 To facilitate more flexible debugging, we also add 4378 console_suspend, a printk module parameter to control 4379 it. Users could use console_suspend (usually 4380 /sys/module/printk/parameters/console_suspend) to 4381 turn on/off it dynamically. 4382 4383 no_debug_objects 4384 [KNL,EARLY] Disable object debugging 4385 4386 nodsp [SH] Disable hardware DSP at boot time. 4387 4388 noefi [EFI,EARLY] Disable EFI runtime services support. 4389 4390 no_entry_flush [PPC,EARLY] Don't flush the L1-D cache when entering the kernel. 4391 4392 noexec32 [X86-64] 4393 This affects only 32-bit executables. 4394 noexec32=on: enable non-executable mappings (default) 4395 read doesn't imply executable mappings 4396 noexec32=off: disable non-executable mappings 4397 read implies executable mappings 4398 4399 no_file_caps Tells the kernel not to honor file capabilities. The 4400 only way then for a file to be executed with privilege 4401 is to be setuid root or executed by root. 4402 4403 nofpu [MIPS,SH] Disable hardware FPU at boot time. 4404 4405 nofsgsbase [X86] Disables FSGSBASE instructions. 4406 4407 nofxsr [BUGS=X86-32] Disables x86 floating point extended 4408 register save and restore. The kernel will only save 4409 legacy floating-point registers on task switch. 4410 4411 nogbpages [X86] Do not use GB pages for kernel direct mappings. 4412 4413 no_hash_pointers 4414 [KNL,EARLY] 4415 Alias for "hash_pointers=never". 4416 4417 nohibernate [HIBERNATION] Disable hibernation and resume. 4418 4419 nohlt [ARM,ARM64,MICROBLAZE,MIPS,PPC,RISCV,SH] Forces the kernel to 4420 busy wait in do_idle() and not use the arch_cpu_idle() 4421 implementation; requires CONFIG_GENERIC_IDLE_POLL_SETUP 4422 to be effective. This is useful on platforms where the 4423 sleep(SH) or wfi(ARM,ARM64) instructions do not work 4424 correctly or when doing power measurements to evaluate 4425 the impact of the sleep instructions. This is also 4426 useful when using JTAG debugger. 4427 4428 nohpet [X86] Don't use the HPET timer. 4429 4430 nohugeiomap [KNL,X86,PPC,ARM64,EARLY] Disable kernel huge I/O mappings. 4431 4432 nohugevmalloc [KNL,X86,PPC,ARM64,EARLY] Disable kernel huge vmalloc mappings. 4433 4434 nohz= [KNL] Boottime enable/disable dynamic ticks 4435 Valid arguments: on, off 4436 Default: on 4437 4438 nohz_full= [KNL,BOOT,SMP,ISOL] 4439 The argument is a cpu list, as described above. 4440 In kernels built with CONFIG_NO_HZ_FULL=y, set 4441 the specified list of CPUs whose tick will be stopped 4442 whenever possible. The boot CPU will be forced outside 4443 the range to maintain the timekeeping. Any CPUs 4444 in this list will have their RCU callbacks offloaded, 4445 just as if they had also been called out in the 4446 rcu_nocbs= boot parameter. 4447 4448 Note that this argument takes precedence over 4449 the CONFIG_RCU_NOCB_CPU_DEFAULT_ALL option. 4450 4451 noinitrd [RAM] Tells the kernel not to load any configured 4452 initial RAM disk. 4453 4454 nointremap [X86-64,Intel-IOMMU,EARLY] Do not enable interrupt 4455 remapping. 4456 [Deprecated - use intremap=off] 4457 4458 noinvpcid [X86,EARLY] Disable the INVPCID cpu feature. 4459 4460 noiotrap [SH] Disables trapped I/O port accesses. 4461 4462 noirqdebug [X86-32] Disables the code which attempts to detect and 4463 disable unhandled interrupt sources. 4464 4465 noisapnp [ISAPNP] Disables ISA PnP code. 4466 4467 nokaslr [KNL,EARLY] 4468 When CONFIG_RANDOMIZE_BASE is set, this disables 4469 kernel and module base offset ASLR (Address Space 4470 Layout Randomization). 4471 4472 no-kvmapf [X86,KVM,EARLY] Disable paravirtualized asynchronous page 4473 fault handling. 4474 4475 no-kvmclock [X86,KVM,EARLY] Disable paravirtualized KVM clock driver 4476 4477 nolapic [X86-32,APIC,EARLY] Do not enable or use the local APIC. 4478 4479 nolapic_timer [X86-32,APIC,EARLY] Do not use the local APIC timer. 4480 4481 nomce [X86-32] Disable Machine Check Exception 4482 4483 nomfgpt [X86-32] Disable Multi-Function General Purpose 4484 Timer usage (for AMD Geode machines). 4485 4486 nomodeset Disable kernel modesetting. Most systems' firmware 4487 sets up a display mode and provides framebuffer memory 4488 for output. With nomodeset, DRM and fbdev drivers will 4489 not load if they could possibly displace the pre- 4490 initialized output. Only the system framebuffer will 4491 be available for use. The respective drivers will not 4492 perform display-mode changes or accelerated rendering. 4493 4494 Useful as error fallback, or for testing and debugging. 4495 4496 nomodule Disable module load 4497 4498 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to 4499 shutdown the other cpus. Instead use the REBOOT_VECTOR 4500 irq. 4501 4502 nopat [X86,EARLY] Disable PAT (page attribute table extension of 4503 pagetables) support. 4504 4505 nopcid [X86-64,EARLY] Disable the PCID cpu feature. 4506 4507 nopku [X86] Disable Memory Protection Keys CPU feature found 4508 in some Intel CPUs. 4509 4510 nopti [X86-64,EARLY] 4511 Equivalent to pti=off 4512 4513 nopv= [X86,XEN,KVM,HYPER_V,VMWARE,EARLY] 4514 Disables the PV optimizations forcing the guest to run 4515 as generic guest with no PV drivers. Currently support 4516 XEN HVM, KVM, HYPER_V and VMWARE guest. 4517 4518 nopvspin [X86,XEN,KVM,EARLY] 4519 Disables the qspinlock slow path using PV optimizations 4520 which allow the hypervisor to 'idle' the guest on lock 4521 contention. 4522 4523 norandmaps Don't use address space randomization. Equivalent to 4524 echo 0 > /proc/sys/kernel/randomize_va_space 4525 4526 noreplace-smp [X86-32,SMP] Don't replace SMP instructions 4527 with UP alternatives 4528 4529 noresume [SWSUSP] Disables resume and restores original swap 4530 space. 4531 4532 no-scroll [VGA] Disables scrollback. 4533 This is required for the Braillex ib80-piezo Braille 4534 reader made by F.H. Papenmeier (Germany). 4535 4536 nosgx [X86-64,SGX,EARLY] Disables Intel SGX kernel support. 4537 4538 nosmap [PPC,EARLY] 4539 Disable SMAP (Supervisor Mode Access Prevention) 4540 even if it is supported by processor. 4541 4542 nosmep [PPC64s,EARLY] 4543 Disable SMEP (Supervisor Mode Execution Prevention) 4544 even if it is supported by processor. 4545 4546 nosmp [SMP,EARLY] Tells an SMP kernel to act as a UP kernel, 4547 and disable the IO APIC. legacy for "maxcpus=0". 4548 4549 nosmt [KNL,MIPS,PPC,EARLY] Disable symmetric multithreading (SMT). 4550 Equivalent to smt=1. 4551 4552 [KNL,X86,PPC,S390] Disable symmetric multithreading (SMT). 4553 nosmt=force: Force disable SMT, cannot be undone 4554 via the sysfs control file. 4555 4556 nosoftlockup [KNL] Disable the soft-lockup detector. 4557 4558 nospec_store_bypass_disable 4559 [HW,EARLY] Disable all mitigations for the Speculative 4560 Store Bypass vulnerability 4561 4562 nospectre_bhb [ARM64,EARLY] Disable all mitigations for Spectre-BHB (branch 4563 history injection) vulnerability. System may allow data leaks 4564 with this option. 4565 4566 nospectre_v1 [X86,PPC,EARLY] Disable mitigations for Spectre Variant 1 4567 (bounds check bypass). With this option data leaks are 4568 possible in the system. 4569 4570 nospectre_v2 [X86,PPC_E500,ARM64,EARLY] Disable all mitigations 4571 for the Spectre variant 2 (indirect branch 4572 prediction) vulnerability. System may allow data 4573 leaks with this option. 4574 4575 no-steal-acc [X86,PV_OPS,ARM64,PPC/PSERIES,RISCV,LOONGARCH,EARLY] 4576 Disable paravirtualized steal time accounting. steal time 4577 is computed, but won't influence scheduler behaviour 4578 4579 nosync [HW,M68K] Disables sync negotiation for all devices. 4580 4581 no_timer_check [X86,APIC] Disables the code which tests for broken 4582 timer IRQ sources, i.e., the IO-APIC timer. This can 4583 work around problems with incorrect timer 4584 initialization on some boards. 4585 4586 no_uaccess_flush 4587 [PPC,EARLY] Don't flush the L1-D cache after accessing user data. 4588 4589 novmcoredd [KNL,KDUMP] 4590 Disable device dump. Device dump allows drivers to 4591 append dump data to vmcore so you can collect driver 4592 specified debug info. Drivers can append the data 4593 without any limit and this data is stored in memory, 4594 so this may cause significant memory stress. Disabling 4595 device dump can help save memory but the driver debug 4596 data will be no longer available. This parameter 4597 is only available when CONFIG_PROC_VMCORE_DEVICE_DUMP 4598 is set. 4599 4600 no-vmw-sched-clock 4601 [X86,PV_OPS,EARLY] Disable paravirtualized VMware 4602 scheduler clock and use the default one. 4603 4604 nowatchdog [KNL] Disable both lockup detectors, i.e. 4605 soft-lockup and NMI watchdog (hard-lockup). 4606 4607 nowb [ARM,EARLY] 4608 4609 nox2apic [X86-64,APIC,EARLY] Do not enable x2APIC mode. 4610 4611 NOTE: this parameter will be ignored on systems with the 4612 LEGACY_XAPIC_DISABLED bit set in the 4613 IA32_XAPIC_DISABLE_STATUS MSR. 4614 4615 noxsave [BUGS=X86] Disables x86 extended register state save 4616 and restore using xsave. The kernel will fallback to 4617 enabling legacy floating-point and sse state. 4618 4619 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended 4620 register states. The kernel will fall back to use 4621 xsave to save the states. By using this parameter, 4622 performance of saving the states is degraded because 4623 xsave doesn't support modified optimization while 4624 xsaveopt supports it on xsaveopt enabled systems. 4625 4626 noxsaves [X86] Disables xsaves and xrstors used in saving and 4627 restoring x86 extended register state in compacted 4628 form of xsave area. The kernel will fall back to use 4629 xsaveopt and xrstor to save and restore the states 4630 in standard form of xsave area. By using this 4631 parameter, xsave area per process might occupy more 4632 memory on xsaves enabled systems. 4633 4634 nr_cpus= [SMP,EARLY] Maximum number of processors that an SMP kernel 4635 could support. nr_cpus=n : n >= 1 limits the kernel to 4636 support 'n' processors. It could be larger than the 4637 number of already plugged CPU during bootup, later in 4638 runtime you can physically add extra cpu until it reaches 4639 n. So during boot up some boot time memory for per-cpu 4640 variables need be pre-allocated for later physical cpu 4641 hot plugging. 4642 4643 nr_uarts= [SERIAL] maximum number of UARTs to be registered. 4644 4645 numa=off [KNL, ARM64, PPC, RISCV, SPARC, X86, EARLY] 4646 Disable NUMA, Only set up a single NUMA node 4647 spanning all memory. 4648 4649 numa=fake=<size>[MG] 4650 [KNL, ARM64, RISCV, X86, EARLY] 4651 If given as a memory unit, fills all system RAM with 4652 nodes of size interleaved over physical nodes. 4653 4654 numa=fake=<N> 4655 [KNL, ARM64, RISCV, X86, EARLY] 4656 If given as an integer, fills all system RAM with N 4657 fake nodes interleaved over physical nodes. 4658 4659 numa=fake=<N>U 4660 [KNL, ARM64, RISCV, X86, EARLY] 4661 If given as an integer followed by 'U', it will 4662 divide each physical node into N emulated nodes. 4663 4664 numa=noacpi [X86] Don't parse the SRAT table for NUMA setup 4665 4666 numa=nohmat [X86] Don't parse the HMAT table for NUMA setup, or 4667 soft-reserved memory partitioning. 4668 4669 numa_balancing= [KNL,ARM64,PPC,RISCV,S390,X86] Enable or disable automatic 4670 NUMA balancing. 4671 Allowed values are enable and disable 4672 4673 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA. 4674 'node', 'default' can be specified 4675 This can be set from sysctl after boot. 4676 See Documentation/admin-guide/sysctl/vm.rst for details. 4677 4678 ohci1394_dma=early [HW,EARLY] enable debugging via the ohci1394 driver. 4679 See Documentation/core-api/debugging-via-ohci1394.rst for more 4680 info. 4681 4682 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands 4683 Rather than timing out after 20 ms if an EC 4684 command is not properly ACKed, override the length 4685 of the timeout. We have interrupts disabled while 4686 waiting for the ACK, so if this is set too high 4687 interrupts *may* be lost! 4688 4689 omap_mux= [OMAP] Override bootloader pin multiplexing. 4690 Format: <mux_mode0.mode_name=value>... 4691 For example, to override I2C bus2: 4692 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100 4693 4694 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration 4695 4696 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock] 4697 4698 boundary - index of last SLC block on Flex-OneNAND. 4699 The remaining blocks are configured as MLC blocks. 4700 lock - Configure if Flex-OneNAND boundary should be locked. 4701 Once locked, the boundary cannot be changed. 4702 1 indicates lock status, 0 indicates unlock status. 4703 4704 oops=panic [KNL,EARLY] 4705 Always panic on oopses. Default is to just kill the 4706 process, but there is a small probability of 4707 deadlocking the machine. 4708 This will also cause panics on machine check exceptions. 4709 Useful together with panic=30 to trigger a reboot. 4710 4711 page_alloc.shuffle= 4712 [KNL] Boolean flag to control whether the page allocator 4713 should randomize its free lists. This parameter can be 4714 used to enable/disable page randomization. The state of 4715 the flag can be read from sysfs at: 4716 /sys/module/page_alloc/parameters/shuffle. 4717 This parameter is only available if CONFIG_SHUFFLE_PAGE_ALLOCATOR=y. 4718 4719 page_owner= [KNL,EARLY] Boot-time page_owner enabling option. 4720 Storage of the information about who allocated 4721 each page is disabled in default. With this switch, 4722 we can turn it on. 4723 on: enable the feature 4724 4725 page_poison= [KNL,EARLY] Boot-time parameter changing the state of 4726 poisoning on the buddy allocator, available with 4727 CONFIG_PAGE_POISONING=y. 4728 off: turn off poisoning (default) 4729 on: turn on poisoning 4730 4731 page_reporting.page_reporting_order= 4732 [KNL] Minimal page reporting order 4733 Format: <integer> 4734 Adjust the minimal page reporting order. The page 4735 reporting is disabled when it exceeds MAX_PAGE_ORDER. 4736 4737 panic= [KNL] Kernel behaviour on panic: delay <timeout> 4738 timeout > 0: seconds before rebooting 4739 timeout = 0: wait forever 4740 timeout < 0: reboot immediately 4741 Format: <timeout> 4742 4743 panic_on_taint= [KNL,EARLY] 4744 Bitmask for conditionally calling panic() in add_taint() 4745 Format: <hex>[,nousertaint] 4746 Hexadecimal bitmask representing the set of TAINT flags 4747 that will cause the kernel to panic when add_taint() is 4748 called with any of the flags in this set. 4749 The optional switch "nousertaint" can be utilized to 4750 prevent userspace forced crashes by writing to sysctl 4751 /proc/sys/kernel/tainted any flagset matching with the 4752 bitmask set on panic_on_taint. 4753 See Documentation/admin-guide/tainted-kernels.rst for 4754 extra details on the taint flags that users can pick 4755 to compose the bitmask to assign to panic_on_taint. 4756 4757 panic_on_warn=1 panic() instead of WARN(). Useful to cause kdump 4758 on a WARN(). 4759 4760 panic_print= Bitmask for printing system info when panic happens. 4761 User can chose combination of the following bits: 4762 bit 0: print all tasks info 4763 bit 1: print system memory info 4764 bit 2: print timer info 4765 bit 3: print locks info if CONFIG_LOCKDEP is on 4766 bit 4: print ftrace buffer 4767 bit 5: replay all kernel messages on consoles at the end of panic 4768 bit 6: print all CPUs backtrace (if available in the arch) 4769 bit 7: print only tasks in uninterruptible (blocked) state 4770 *Be aware* that this option may print a _lot_ of lines, 4771 so there are risks of losing older messages in the log. 4772 Use this option carefully, maybe worth to setup a 4773 bigger log buffer with "log_buf_len" along with this. 4774 4775 panic_sys_info= A comma separated list of extra information to be dumped 4776 on panic. 4777 Format: val[,val...] 4778 Where @val can be any of the following: 4779 4780 tasks: print all tasks info 4781 mem: print system memory info 4782 timers: print timers info 4783 locks: print locks info if CONFIG_LOCKDEP is on 4784 ftrace: print ftrace buffer 4785 all_bt: print all CPUs backtrace (if available in the arch) 4786 blocked_tasks: print only tasks in uninterruptible (blocked) state 4787 4788 This is a human readable alternative to the 'panic_print' option. 4789 4790 panic_console_replay 4791 When panic happens, replay all kernel messages on 4792 consoles at the end of panic. 4793 4794 parkbd.port= [HW] Parallel port number the keyboard adapter is 4795 connected to, default is 0. 4796 Format: <parport#> 4797 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation, 4798 0 for XT, 1 for AT (default is AT). 4799 Format: <mode> 4800 4801 parport= [HW,PPT] Specify parallel ports. 0 disables. 4802 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] } 4803 Use 'auto' to force the driver to use any 4804 IRQ/DMA settings detected (the default is to 4805 ignore detected IRQ/DMA settings because of 4806 possible conflicts). You can specify the base 4807 address, IRQ, and DMA settings; IRQ and DMA 4808 should be numbers, or 'auto' (for using detected 4809 settings on that particular port), or 'nofifo' 4810 (to avoid using a FIFO even if it is detected). 4811 Parallel ports are assigned in the order they 4812 are specified on the command line, starting 4813 with parport0. 4814 4815 parport_init_mode= [HW,PPT] 4816 Configure VIA parallel port to operate in 4817 a specific mode. This is necessary on Pegasos 4818 computer where firmware has no options for setting 4819 up parallel port mode and sets it to spp. 4820 Currently this function knows 686a and 8231 chips. 4821 Format: [spp|ps2|epp|ecp|ecpepp] 4822 4823 pata_legacy.all= [HW,LIBATA] 4824 Format: <int> 4825 Set to non-zero to probe primary and secondary ISA 4826 port ranges on PCI systems where no PCI PATA device 4827 has been found at either range. Disabled by default. 4828 4829 pata_legacy.autospeed= [HW,LIBATA] 4830 Format: <int> 4831 Set to non-zero if a chip is present that snoops speed 4832 changes. Disabled by default. 4833 4834 pata_legacy.ht6560a= [HW,LIBATA] 4835 Format: <int> 4836 Set to 1, 2, or 3 for HT 6560A on the primary channel, 4837 the secondary channel, or both channels respectively. 4838 Disabled by default. 4839 4840 pata_legacy.ht6560b= [HW,LIBATA] 4841 Format: <int> 4842 Set to 1, 2, or 3 for HT 6560B on the primary channel, 4843 the secondary channel, or both channels respectively. 4844 Disabled by default. 4845 4846 pata_legacy.iordy_mask= [HW,LIBATA] 4847 Format: <int> 4848 IORDY enable mask. Set individual bits to allow IORDY 4849 for the respective channel. Bit 0 is for the first 4850 legacy channel handled by this driver, bit 1 is for 4851 the second channel, and so on. The sequence will often 4852 correspond to the primary legacy channel, the secondary 4853 legacy channel, and so on, but the handling of a PCI 4854 bus and the use of other driver options may interfere 4855 with the sequence. By default IORDY is allowed across 4856 all channels. 4857 4858 pata_legacy.opti82c46x= [HW,LIBATA] 4859 Format: <int> 4860 Set to 1, 2, or 3 for Opti 82c611A on the primary 4861 channel, the secondary channel, or both channels 4862 respectively. Disabled by default. 4863 4864 pata_legacy.opti82c611a= [HW,LIBATA] 4865 Format: <int> 4866 Set to 1, 2, or 3 for Opti 82c465MV on the primary 4867 channel, the secondary channel, or both channels 4868 respectively. Disabled by default. 4869 4870 pata_legacy.pio_mask= [HW,LIBATA] 4871 Format: <int> 4872 PIO mode mask for autospeed devices. Set individual 4873 bits to allow the use of the respective PIO modes. 4874 Bit 0 is for mode 0, bit 1 is for mode 1, and so on. 4875 All modes allowed by default. 4876 4877 pata_legacy.probe_all= [HW,LIBATA] 4878 Format: <int> 4879 Set to non-zero to probe tertiary and further ISA 4880 port ranges on PCI systems. Disabled by default. 4881 4882 pata_legacy.probe_mask= [HW,LIBATA] 4883 Format: <int> 4884 Probe mask for legacy ISA PATA ports. Depending on 4885 platform configuration and the use of other driver 4886 options up to 6 legacy ports are supported: 0x1f0, 4887 0x170, 0x1e8, 0x168, 0x1e0, 0x160, however probing 4888 of individual ports can be disabled by setting the 4889 corresponding bits in the mask to 1. Bit 0 is for 4890 the first port in the list above (0x1f0), and so on. 4891 By default all supported ports are probed. 4892 4893 pata_legacy.qdi= [HW,LIBATA] 4894 Format: <int> 4895 Set to non-zero to probe QDI controllers. By default 4896 set to 1 if CONFIG_PATA_QDI_MODULE, 0 otherwise. 4897 4898 pata_legacy.winbond= [HW,LIBATA] 4899 Format: <int> 4900 Set to non-zero to probe Winbond controllers. Use 4901 the standard I/O port (0x130) if 1, otherwise the 4902 value given is the I/O port to use (typically 0x1b0). 4903 By default set to 1 if CONFIG_PATA_WINBOND_VLB_MODULE, 4904 0 otherwise. 4905 4906 pata_platform.pio_mask= [HW,LIBATA] 4907 Format: <int> 4908 Supported PIO mode mask. Set individual bits to allow 4909 the use of the respective PIO modes. Bit 0 is for 4910 mode 0, bit 1 is for mode 1, and so on. Mode 0 only 4911 allowed by default. 4912 4913 pause_on_oops=<int> 4914 Halt all CPUs after the first oops has been printed for 4915 the specified number of seconds. This is to be used if 4916 your oopses keep scrolling off the screen. 4917 4918 pcbit= [HW,ISDN] 4919 4920 pci=option[,option...] [PCI,EARLY] various PCI subsystem options. 4921 4922 Some options herein operate on a specific device 4923 or a set of devices (<pci_dev>). These are 4924 specified in one of the following formats: 4925 4926 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]* 4927 pci:<vendor>:<device>[:<subvendor>:<subdevice>] 4928 4929 Note: the first format specifies a PCI 4930 bus/device/function address which may change 4931 if new hardware is inserted, if motherboard 4932 firmware changes, or due to changes caused 4933 by other kernel parameters. If the 4934 domain is left unspecified, it is 4935 taken to be zero. Optionally, a path 4936 to a device through multiple device/function 4937 addresses can be specified after the base 4938 address (this is more robust against 4939 renumbering issues). The second format 4940 selects devices using IDs from the 4941 configuration space which may match multiple 4942 devices in the system. 4943 4944 earlydump dump PCI config space before the kernel 4945 changes anything 4946 off [X86] don't probe for the PCI bus 4947 bios [X86-32] force use of PCI BIOS, don't access 4948 the hardware directly. Use this if your machine 4949 has a non-standard PCI host bridge. 4950 nobios [X86-32] disallow use of PCI BIOS, only direct 4951 hardware access methods are allowed. Use this 4952 if you experience crashes upon bootup and you 4953 suspect they are caused by the BIOS. 4954 conf1 [X86] Force use of PCI Configuration Access 4955 Mechanism 1 (config address in IO port 0xCF8, 4956 data in IO port 0xCFC, both 32-bit). 4957 conf2 [X86] Force use of PCI Configuration Access 4958 Mechanism 2 (IO port 0xCF8 is an 8-bit port for 4959 the function, IO port 0xCFA, also 8-bit, sets 4960 bus number. The config space is then accessed 4961 through ports 0xC000-0xCFFF). 4962 See http://wiki.osdev.org/PCI for more info 4963 on the configuration access mechanisms. 4964 noaer [PCIE] If the PCIEAER kernel config parameter is 4965 enabled, this kernel boot option can be used to 4966 disable the use of PCIE advanced error reporting. 4967 nodomains [PCI] Disable support for multiple PCI 4968 root domains (aka PCI segments, in ACPI-speak). 4969 nommconf [X86] Disable use of MMCONFIG for PCI 4970 Configuration 4971 check_enable_amd_mmconf [X86] check for and enable 4972 properly configured MMIO access to PCI 4973 config space on AMD family 10h CPU 4974 nomsi [MSI] If the PCI_MSI kernel config parameter is 4975 enabled, this kernel boot option can be used to 4976 disable the use of MSI interrupts system-wide. 4977 noioapicquirk [APIC] Disable all boot interrupt quirks. 4978 Safety option to keep boot IRQs enabled. This 4979 should never be necessary. 4980 ioapicreroute [APIC] Enable rerouting of boot IRQs to the 4981 primary IO-APIC for bridges that cannot disable 4982 boot IRQs. This fixes a source of spurious IRQs 4983 when the system masks IRQs. 4984 noioapicreroute [APIC] Disable workaround that uses the 4985 boot IRQ equivalent of an IRQ that connects to 4986 a chipset where boot IRQs cannot be disabled. 4987 The opposite of ioapicreroute. 4988 biosirq [X86-32] Use PCI BIOS calls to get the interrupt 4989 routing table. These calls are known to be buggy 4990 on several machines and they hang the machine 4991 when used, but on other computers it's the only 4992 way to get the interrupt routing table. Try 4993 this option if the kernel is unable to allocate 4994 IRQs or discover secondary PCI buses on your 4995 motherboard. 4996 rom [X86] Assign address space to expansion ROMs. 4997 Use with caution as certain devices share 4998 address decoders between ROMs and other 4999 resources. 5000 norom [X86] Do not assign address space to 5001 expansion ROMs that do not already have 5002 BIOS assigned address ranges. 5003 nobar [X86] Do not assign address space to the 5004 BARs that weren't assigned by the BIOS. 5005 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be 5006 assigned automatically to PCI devices. You can 5007 make the kernel exclude IRQs of your ISA cards 5008 this way. 5009 pirqaddr=0xAAAAA [X86] Specify the physical address 5010 of the PIRQ table (normally generated 5011 by the BIOS) if it is outside the 5012 F0000h-100000h range. 5013 lastbus=N [X86] Scan all buses thru bus #N. Can be 5014 useful if the kernel is unable to find your 5015 secondary buses and you want to tell it 5016 explicitly which ones they are. 5017 assign-busses [X86] Always assign all PCI bus 5018 numbers ourselves, overriding 5019 whatever the firmware may have done. 5020 usepirqmask [X86] Honor the possible IRQ mask stored 5021 in the BIOS $PIR table. This is needed on 5022 some systems with broken BIOSes, notably 5023 some HP Pavilion N5400 and Omnibook XE3 5024 notebooks. This will have no effect if ACPI 5025 IRQ routing is enabled. 5026 noacpi [X86] Do not use ACPI for IRQ routing 5027 or for PCI scanning. 5028 use_crs [X86] Use PCI host bridge window information 5029 from ACPI. On BIOSes from 2008 or later, this 5030 is enabled by default. If you need to use this, 5031 please report a bug. 5032 nocrs [X86] Ignore PCI host bridge windows from ACPI. 5033 If you need to use this, please report a bug. 5034 use_e820 [X86] Use E820 reservations to exclude parts of 5035 PCI host bridge windows. This is a workaround 5036 for BIOS defects in host bridge _CRS methods. 5037 If you need to use this, please report a bug to 5038 <linux-pci@vger.kernel.org>. 5039 no_e820 [X86] Ignore E820 reservations for PCI host 5040 bridge windows. This is the default on modern 5041 hardware. If you need to use this, please report 5042 a bug to <linux-pci@vger.kernel.org>. 5043 routeirq Do IRQ routing for all PCI devices. 5044 This is normally done in pci_enable_device(), 5045 so this option is a temporary workaround 5046 for broken drivers that don't call it. 5047 skip_isa_align [X86] do not align io start addr, so can 5048 handle more pci cards 5049 noearly [X86] Don't do any early type 1 scanning. 5050 This might help on some broken boards which 5051 machine check when some devices' config space 5052 is read. But various workarounds are disabled 5053 and some IOMMU drivers will not work. 5054 bfsort Sort PCI devices into breadth-first order. 5055 This sorting is done to get a device 5056 order compatible with older (<= 2.4) kernels. 5057 nobfsort Don't sort PCI devices into breadth-first order. 5058 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size) 5059 tuning and use the BIOS-configured MPS defaults. 5060 pcie_bus_safe Set every device's MPS to the largest value 5061 supported by all devices below the root complex. 5062 pcie_bus_perf Set device MPS to the largest allowable MPS 5063 based on its parent bus. Also set MRRS (Max 5064 Read Request Size) to the largest supported 5065 value (no larger than the MPS that the device 5066 or bus can support) for best performance. 5067 pcie_bus_peer2peer Set every device's MPS to 128B, which 5068 every device is guaranteed to support. This 5069 configuration allows peer-to-peer DMA between 5070 any pair of devices, possibly at the cost of 5071 reduced performance. This also guarantees 5072 that hot-added devices will work. 5073 cbiosize=nn[KMG] The fixed amount of bus space which is 5074 reserved for the CardBus bridge's IO window. 5075 The default value is 256 bytes. 5076 cbmemsize=nn[KMG] The fixed amount of bus space which is 5077 reserved for the CardBus bridge's memory 5078 window. The default value is 64 megabytes. 5079 resource_alignment= 5080 Format: 5081 [<order of align>@]<pci_dev>[; ...] 5082 Specifies alignment and device to reassign 5083 aligned memory resources. How to 5084 specify the device is described above. 5085 If <order of align> is not specified, 5086 PAGE_SIZE is used as alignment. 5087 A PCI-PCI bridge can be specified if resource 5088 windows need to be expanded. 5089 To specify the alignment for several 5090 instances of a device, the PCI vendor, 5091 device, subvendor, and subdevice may be 5092 specified, e.g., 12@pci:8086:9c22:103c:198f 5093 for 4096-byte alignment. 5094 ecrc= Enable/disable PCIe ECRC (transaction layer 5095 end-to-end CRC checking). Only effective if 5096 OS has native AER control (either granted by 5097 ACPI _OSC or forced via "pcie_ports=native") 5098 bios: Use BIOS/firmware settings. This is the 5099 the default. 5100 off: Turn ECRC off 5101 on: Turn ECRC on. 5102 hpiosize=nn[KMG] The fixed amount of bus space which is 5103 reserved for hotplug bridge's IO window. 5104 Default size is 256 bytes. 5105 hpmmiosize=nn[KMG] The fixed amount of bus space which is 5106 reserved for hotplug bridge's MMIO window. 5107 Default size is 2 megabytes. 5108 hpmmioprefsize=nn[KMG] The fixed amount of bus space which is 5109 reserved for hotplug bridge's MMIO_PREF window. 5110 Default size is 2 megabytes. 5111 hpmemsize=nn[KMG] The fixed amount of bus space which is 5112 reserved for hotplug bridge's MMIO and 5113 MMIO_PREF window. 5114 Default size is 2 megabytes. 5115 hpbussize=nn The minimum amount of additional bus numbers 5116 reserved for buses below a hotplug bridge. 5117 Default is 1. 5118 realloc= Enable/disable reallocating PCI bridge resources 5119 if allocations done by BIOS are too small to 5120 accommodate resources required by all child 5121 devices. 5122 off: Turn realloc off 5123 on: Turn realloc on 5124 realloc same as realloc=on 5125 noari do not use PCIe ARI. 5126 noats [PCIE, Intel-IOMMU, AMD-IOMMU] 5127 do not use PCIe ATS (and IOMMU device IOTLB). 5128 pcie_scan_all Scan all possible PCIe devices. Otherwise we 5129 only look for one device below a PCIe downstream 5130 port. 5131 big_root_window Try to add a big 64bit memory window to the PCIe 5132 root complex on AMD CPUs. Some GFX hardware 5133 can resize a BAR to allow access to all VRAM. 5134 Adding the window is slightly risky (it may 5135 conflict with unreported devices), so this 5136 taints the kernel. 5137 disable_acs_redir=<pci_dev>[; ...] 5138 Specify one or more PCI devices (in the format 5139 specified above) separated by semicolons. 5140 Each device specified will have the PCI ACS 5141 redirect capabilities forced off which will 5142 allow P2P traffic between devices through 5143 bridges without forcing it upstream. Note: 5144 this removes isolation between devices and 5145 may put more devices in an IOMMU group. 5146 config_acs= 5147 Format: 5148 <ACS flags>@<pci_dev>[; ...] 5149 Specify one or more PCI devices (in the format 5150 specified above) optionally prepended with flags 5151 and separated by semicolons. The respective 5152 capabilities will be enabled, disabled or 5153 unchanged based on what is specified in 5154 flags. 5155 5156 ACS Flags is defined as follows: 5157 bit-0 : ACS Source Validation 5158 bit-1 : ACS Translation Blocking 5159 bit-2 : ACS P2P Request Redirect 5160 bit-3 : ACS P2P Completion Redirect 5161 bit-4 : ACS Upstream Forwarding 5162 bit-5 : ACS P2P Egress Control 5163 bit-6 : ACS Direct Translated P2P 5164 Each bit can be marked as: 5165 '0' – force disabled 5166 '1' – force enabled 5167 'x' – unchanged 5168 For example, 5169 pci=config_acs=10x@pci:0:0 5170 would configure all devices that support 5171 ACS to enable P2P Request Redirect, disable 5172 Translation Blocking, and leave Source 5173 Validation unchanged from whatever power-up 5174 or firmware set it to. 5175 5176 Note: this may remove isolation between devices 5177 and may put more devices in an IOMMU group. 5178 force_floating [S390] Force usage of floating interrupts. 5179 nomio [S390] Do not use MIO instructions. 5180 norid [S390] ignore the RID field and force use of 5181 one PCI domain per PCI function 5182 notph [PCIE] If the PCIE_TPH kernel config parameter 5183 is enabled, this kernel boot option can be used 5184 to disable PCIe TLP Processing Hints support 5185 system-wide. 5186 5187 pcie_aspm= [PCIE] Forcibly enable or ignore PCIe Active State Power 5188 Management. 5189 off Don't touch ASPM configuration at all. Leave any 5190 configuration done by firmware unchanged. 5191 force Enable ASPM even on devices that claim not to support it. 5192 WARNING: Forcing ASPM on may cause system lockups. 5193 5194 pcie_ports= [PCIE] PCIe port services handling: 5195 native Use native PCIe services (PME, AER, DPC, PCIe hotplug) 5196 even if the platform doesn't give the OS permission to 5197 use them. This may cause conflicts if the platform 5198 also tries to use these services. 5199 dpc-native Use native PCIe service for DPC only. May 5200 cause conflicts if firmware uses AER or DPC. 5201 compat Disable native PCIe services (PME, AER, DPC, PCIe 5202 hotplug). 5203 5204 pcie_port_pm= [PCIE] PCIe port power management handling: 5205 off Disable power management of all PCIe ports 5206 force Forcibly enable power management of all PCIe ports 5207 5208 pcie_pme= [PCIE,PM] Native PCIe PME signaling options: 5209 nomsi Do not use MSI for native PCIe PME signaling (this makes 5210 all PCIe root ports use INTx for all services). 5211 5212 pcmv= [HW,PCMCIA] BadgePAD 4 5213 5214 pd_ignore_unused 5215 [PM] 5216 Keep all power-domains already enabled by bootloader on, 5217 even if no driver has claimed them. This is useful 5218 for debug and development, but should not be 5219 needed on a platform with proper driver support. 5220 5221 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at 5222 boot time. 5223 Format: { 0 | 1 } 5224 See arch/parisc/kernel/pdc_chassis.c 5225 5226 percpu_alloc= [MM,EARLY] 5227 Select which percpu first chunk allocator to use. 5228 Currently supported values are "embed" and "page". 5229 Archs may support subset or none of the selections. 5230 See comments in mm/percpu.c for details on each 5231 allocator. This parameter is primarily for debugging 5232 and performance comparison. 5233 5234 pirq= [SMP,APIC] Manual mp-table setup 5235 See Documentation/arch/x86/i386/IO-APIC.rst. 5236 5237 plip= [PPT,NET] Parallel port network link 5238 Format: { parport<nr> | timid | 0 } 5239 See also Documentation/admin-guide/parport.rst. 5240 5241 pmtmr= [X86] Manual setup of pmtmr I/O Port. 5242 Override pmtimer IOPort with a hex value. 5243 e.g. pmtmr=0x508 5244 5245 pmu_override= [PPC] Override the PMU. 5246 This option takes over the PMU facility, so it is no 5247 longer usable by perf. Setting this option starts the 5248 PMU counters by setting MMCR0 to 0 (the FC bit is 5249 cleared). If a number is given, then MMCR1 is set to 5250 that number, otherwise (e.g., 'pmu_override=on'), MMCR1 5251 remains 0. 5252 5253 pm_async= [PM] 5254 Format: off 5255 This parameter sets the initial value of the 5256 /sys/power/pm_async sysfs knob at boot time. 5257 If set to "off", disables asynchronous suspend and 5258 resume of devices during system-wide power transitions. 5259 This can be useful on platforms where device 5260 dependencies are not well-defined, or for debugging 5261 power management issues. Asynchronous operations are 5262 enabled by default. 5263 5264 5265 pm_debug_messages [SUSPEND,KNL] 5266 Enable suspend/resume debug messages during boot up. 5267 5268 pnp.debug=1 [PNP] 5269 Enable PNP debug messages (depends on the 5270 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time 5271 via /sys/module/pnp/parameters/debug. We always show 5272 current resource usage; turning this on also shows 5273 possible settings and some assignment information. 5274 5275 pnpacpi= [ACPI] 5276 { off } 5277 5278 pnpbios= [ISAPNP] 5279 { on | off | curr | res | no-curr | no-res } 5280 5281 pnp_reserve_irq= 5282 [ISAPNP] Exclude IRQs for the autoconfiguration 5283 5284 pnp_reserve_dma= 5285 [ISAPNP] Exclude DMAs for the autoconfiguration 5286 5287 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration 5288 Ranges are in pairs (I/O port base and size). 5289 5290 pnp_reserve_mem= 5291 [ISAPNP] Exclude memory regions for the 5292 autoconfiguration. 5293 Ranges are in pairs (memory base and size). 5294 5295 ports= [IP_VS_FTP] IPVS ftp helper module 5296 Default is 21. 5297 Up to 8 (IP_VS_APP_MAX_PORTS) ports 5298 may be specified. 5299 Format: <port>,<port>.... 5300 5301 possible_cpus= [SMP,S390,X86] 5302 Format: <unsigned int> 5303 Set the number of possible CPUs, overriding the 5304 regular discovery mechanisms (such as ACPI/FW, etc). 5305 5306 powersave=off [PPC] This option disables power saving features. 5307 It specifically disables cpuidle and sets the 5308 platform machine description specific power_save 5309 function to NULL. On Idle the CPU just reduces 5310 execution priority. 5311 5312 ppc_strict_facility_enable 5313 [PPC,ENABLE] This option catches any kernel floating point, 5314 Altivec, VSX and SPE outside of regions specifically 5315 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()). 5316 There is some performance impact when enabling this. 5317 5318 ppc_tm= [PPC,EARLY] 5319 Format: {"off"} 5320 Disable Hardware Transactional Memory 5321 5322 preempt= [KNL] 5323 Select preemption mode if you have CONFIG_PREEMPT_DYNAMIC 5324 none - Limited to cond_resched() calls 5325 voluntary - Limited to cond_resched() and might_sleep() calls 5326 full - Any section that isn't explicitly preempt disabled 5327 can be preempted anytime. Tasks will also yield 5328 contended spinlocks (if the critical section isn't 5329 explicitly preempt disabled beyond the lock itself). 5330 lazy - Scheduler controlled. Similar to full but instead 5331 of preempting the task immediately, the task gets 5332 one HZ tick time to yield itself before the 5333 preemption will be forced. One preemption is when the 5334 task returns to user space. 5335 5336 print-fatal-signals= 5337 [KNL] debug: print fatal signals 5338 5339 If enabled, warn about various signal handling 5340 related application anomalies: too many signals, 5341 too many POSIX.1 timers, fatal signals causing a 5342 coredump - etc. 5343 5344 If you hit the warning due to signal overflow, 5345 you might want to try "ulimit -i unlimited". 5346 5347 default: off. 5348 5349 printk.always_kmsg_dump= 5350 Trigger kmsg_dump for cases other than kernel oops or 5351 panics 5352 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 5353 default: disabled 5354 5355 printk.console_no_auto_verbose= 5356 Disable console loglevel raise on oops, panic 5357 or lockdep-detected issues (only if lock debug is on). 5358 With an exception to setups with low baudrate on 5359 serial console, keeping this 0 is a good choice 5360 in order to provide more debug information. 5361 Format: <bool> 5362 default: 0 (auto_verbose is enabled) 5363 5364 printk.debug_non_panic_cpus= 5365 Allows storing messages from non-panic CPUs into 5366 the printk log buffer during panic(). They are 5367 flushed to consoles by the panic-CPU on 5368 a best-effort basis. 5369 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 5370 Default: disabled 5371 5372 printk.devkmsg={on,off,ratelimit} 5373 Control writing to /dev/kmsg. 5374 on - unlimited logging to /dev/kmsg from userspace 5375 off - logging to /dev/kmsg disabled 5376 ratelimit - ratelimit the logging 5377 Default: ratelimit 5378 5379 printk.time= Show timing data prefixed to each printk message line 5380 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 5381 5382 proc_mem.force_override= [KNL] 5383 Format: {always | ptrace | never} 5384 Traditionally /proc/pid/mem allows memory permissions to be 5385 overridden without restrictions. This option may be set to 5386 restrict that. Can be one of: 5387 - 'always': traditional behavior always allows mem overrides. 5388 - 'ptrace': only allow mem overrides for active ptracers. 5389 - 'never': never allow mem overrides. 5390 If not specified, default is the CONFIG_PROC_MEM_* choice. 5391 5392 processor.max_cstate= [HW,ACPI] 5393 Limit processor to maximum C-state 5394 max_cstate=9 overrides any DMI blacklist limit. 5395 5396 processor.nocst [HW,ACPI] 5397 Ignore the _CST method to determine C-states, 5398 instead using the legacy FADT method 5399 5400 profile= [KNL] Enable kernel profiling via /proc/profile 5401 Format: [<profiletype>,]<number> 5402 Param: <profiletype>: "schedule" or "kvm" 5403 [defaults to kernel profiling] 5404 Param: "schedule" - profile schedule points. 5405 Param: "kvm" - profile VM exits. 5406 Param: <number> - step/bucket size as a power of 2 for 5407 statistical time based profiling. 5408 5409 prompt_ramdisk= [RAM] [Deprecated] 5410 5411 prot_virt= [S390] enable hosting protected virtual machines 5412 isolated from the hypervisor (if hardware supports 5413 that). If enabled, the default kernel base address 5414 might be overridden even when Kernel Address Space 5415 Layout Randomization is disabled. 5416 Format: <bool> 5417 5418 psi= [KNL] Enable or disable pressure stall information 5419 tracking. 5420 Format: <bool> 5421 5422 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to 5423 probe for; one of (bare|imps|exps|lifebook|any). 5424 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports 5425 per second. 5426 psmouse.resetafter= [HW,MOUSE] 5427 Try to reset the device after so many bad packets 5428 (0 = never). 5429 psmouse.resolution= 5430 [HW,MOUSE] Set desired mouse resolution, in dpi. 5431 psmouse.smartscroll= 5432 [HW,MOUSE] Controls Logitech smartscroll autorepeat. 5433 0 = disabled, 1 = enabled (default). 5434 5435 pstore.backend= Specify the name of the pstore backend to use 5436 5437 pti= [X86-64] Control Page Table Isolation of user and 5438 kernel address spaces. Disabling this feature 5439 removes hardening, but improves performance of 5440 system calls and interrupts. 5441 5442 on - unconditionally enable 5443 off - unconditionally disable 5444 auto - kernel detects whether your CPU model is 5445 vulnerable to issues that PTI mitigates 5446 5447 Not specifying this option is equivalent to pti=auto. 5448 5449 pty.legacy_count= 5450 [KNL] Number of legacy pty's. Overwrites compiled-in 5451 default number. 5452 5453 quiet [KNL,EARLY] Disable most log messages 5454 5455 r128= [HW,DRM] 5456 5457 radix_hcall_invalidate=on [PPC/PSERIES] 5458 Disable RADIX GTSE feature and use hcall for TLB 5459 invalidate. 5460 5461 raid= [HW,RAID] 5462 See Documentation/admin-guide/md.rst. 5463 5464 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes 5465 See Documentation/admin-guide/blockdev/ramdisk.rst. 5466 5467 ramdisk_start= [RAM] RAM disk image start address 5468 5469 random.trust_cpu=off 5470 [KNL,EARLY] Disable trusting the use of the CPU's 5471 random number generator (if available) to 5472 initialize the kernel's RNG. 5473 5474 random.trust_bootloader=off 5475 [KNL,EARLY] Disable trusting the use of the a seed 5476 passed by the bootloader (if available) to 5477 initialize the kernel's RNG. 5478 5479 randomize_kstack_offset= 5480 [KNL,EARLY] Enable or disable kernel stack offset 5481 randomization, which provides roughly 5 bits of 5482 entropy, frustrating memory corruption attacks 5483 that depend on stack address determinism or 5484 cross-syscall address exposures. This is only 5485 available on architectures that have defined 5486 CONFIG_HAVE_ARCH_RANDOMIZE_KSTACK_OFFSET. 5487 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 5488 Default is CONFIG_RANDOMIZE_KSTACK_OFFSET_DEFAULT. 5489 5490 ras=option[,option,...] [KNL] RAS-specific options 5491 5492 cec_disable [X86] 5493 Disable the Correctable Errors Collector, 5494 see CONFIG_RAS_CEC help text. 5495 5496 rcu_nocbs[=cpu-list] 5497 [KNL] The optional argument is a cpu list, 5498 as described above. 5499 5500 In kernels built with CONFIG_RCU_NOCB_CPU=y, 5501 enable the no-callback CPU mode, which prevents 5502 such CPUs' callbacks from being invoked in 5503 softirq context. Invocation of such CPUs' RCU 5504 callbacks will instead be offloaded to "rcuox/N" 5505 kthreads created for that purpose, where "x" is 5506 "p" for RCU-preempt, "s" for RCU-sched, and "g" 5507 for the kthreads that mediate grace periods; and 5508 "N" is the CPU number. This reduces OS jitter on 5509 the offloaded CPUs, which can be useful for HPC 5510 and real-time workloads. It can also improve 5511 energy efficiency for asymmetric multiprocessors. 5512 5513 If a cpulist is passed as an argument, the specified 5514 list of CPUs is set to no-callback mode from boot. 5515 5516 Otherwise, if the '=' sign and the cpulist 5517 arguments are omitted, no CPU will be set to 5518 no-callback mode from boot but the mode may be 5519 toggled at runtime via cpusets. 5520 5521 Note that this argument takes precedence over 5522 the CONFIG_RCU_NOCB_CPU_DEFAULT_ALL option. 5523 5524 rcu_nocb_poll [KNL] 5525 Rather than requiring that offloaded CPUs 5526 (specified by rcu_nocbs= above) explicitly 5527 awaken the corresponding "rcuoN" kthreads, 5528 make these kthreads poll for callbacks. 5529 This improves the real-time response for the 5530 offloaded CPUs by relieving them of the need to 5531 wake up the corresponding kthread, but degrades 5532 energy efficiency by requiring that the kthreads 5533 periodically wake up to do the polling. 5534 5535 rcutree.blimit= [KNL] 5536 Set maximum number of finished RCU callbacks to 5537 process in one batch. 5538 5539 rcutree.csd_lock_suppress_rcu_stall= [KNL] 5540 Do only a one-line RCU CPU stall warning when 5541 there is an ongoing too-long CSD-lock wait. 5542 5543 rcutree.do_rcu_barrier= [KNL] 5544 Request a call to rcu_barrier(). This is 5545 throttled so that userspace tests can safely 5546 hammer on the sysfs variable if they so choose. 5547 If triggered before the RCU grace-period machinery 5548 is fully active, this will error out with EAGAIN. 5549 5550 rcutree.dump_tree= [KNL] 5551 Dump the structure of the rcu_node combining tree 5552 out at early boot. This is used for diagnostic 5553 purposes, to verify correct tree setup. 5554 5555 rcutree.gp_cleanup_delay= [KNL] 5556 Set the number of jiffies to delay each step of 5557 RCU grace-period cleanup. 5558 5559 rcutree.gp_init_delay= [KNL] 5560 Set the number of jiffies to delay each step of 5561 RCU grace-period initialization. 5562 5563 rcutree.gp_preinit_delay= [KNL] 5564 Set the number of jiffies to delay each step of 5565 RCU grace-period pre-initialization, that is, 5566 the propagation of recent CPU-hotplug changes up 5567 the rcu_node combining tree. 5568 5569 rcutree.jiffies_till_first_fqs= [KNL] 5570 Set delay from grace-period initialization to 5571 first attempt to force quiescent states. 5572 Units are jiffies, minimum value is zero, 5573 and maximum value is HZ. 5574 5575 rcutree.jiffies_till_next_fqs= [KNL] 5576 Set delay between subsequent attempts to force 5577 quiescent states. Units are jiffies, minimum 5578 value is one, and maximum value is HZ. 5579 5580 rcutree.jiffies_till_sched_qs= [KNL] 5581 Set required age in jiffies for a 5582 given grace period before RCU starts 5583 soliciting quiescent-state help from 5584 rcu_note_context_switch() and cond_resched(). 5585 If not specified, the kernel will calculate 5586 a value based on the most recent settings 5587 of rcutree.jiffies_till_first_fqs 5588 and rcutree.jiffies_till_next_fqs. 5589 This calculated value may be viewed in 5590 rcutree.jiffies_to_sched_qs. Any attempt to set 5591 rcutree.jiffies_to_sched_qs will be cheerfully 5592 overwritten. 5593 5594 rcutree.kthread_prio= [KNL,BOOT] 5595 Set the SCHED_FIFO priority of the RCU per-CPU 5596 kthreads (rcuc/N). This value is also used for 5597 the priority of the RCU boost threads (rcub/N) 5598 and for the RCU grace-period kthreads (rcu_bh, 5599 rcu_preempt, and rcu_sched). If RCU_BOOST is 5600 set, valid values are 1-99 and the default is 1 5601 (the least-favored priority). Otherwise, when 5602 RCU_BOOST is not set, valid values are 0-99 and 5603 the default is zero (non-realtime operation). 5604 When RCU_NOCB_CPU is set, also adjust the 5605 priority of NOCB callback kthreads. 5606 5607 rcutree.nocb_nobypass_lim_per_jiffy= [KNL] 5608 On callback-offloaded (rcu_nocbs) CPUs, 5609 RCU reduces the lock contention that would 5610 otherwise be caused by callback floods through 5611 use of the ->nocb_bypass list. However, in the 5612 common non-flooded case, RCU queues directly to 5613 the main ->cblist in order to avoid the extra 5614 overhead of the ->nocb_bypass list and its lock. 5615 But if there are too many callbacks queued during 5616 a single jiffy, RCU pre-queues the callbacks into 5617 the ->nocb_bypass queue. The definition of "too 5618 many" is supplied by this kernel boot parameter. 5619 5620 rcutree.nohz_full_patience_delay= [KNL] 5621 On callback-offloaded (rcu_nocbs) CPUs, avoid 5622 disturbing RCU unless the grace period has 5623 reached the specified age in milliseconds. 5624 Defaults to zero. Large values will be capped 5625 at five seconds. All values will be rounded down 5626 to the nearest value representable by jiffies. 5627 5628 rcutree.qhimark= [KNL] 5629 Set threshold of queued RCU callbacks beyond which 5630 batch limiting is disabled. 5631 5632 rcutree.qlowmark= [KNL] 5633 Set threshold of queued RCU callbacks below which 5634 batch limiting is re-enabled. 5635 5636 rcutree.qovld= [KNL] 5637 Set threshold of queued RCU callbacks beyond which 5638 RCU's force-quiescent-state scan will aggressively 5639 enlist help from cond_resched() and sched IPIs to 5640 help CPUs more quickly reach quiescent states. 5641 Set to less than zero to make this be set based 5642 on rcutree.qhimark at boot time and to zero to 5643 disable more aggressive help enlistment. 5644 5645 rcutree.rcu_delay_page_cache_fill_msec= [KNL] 5646 Set the page-cache refill delay (in milliseconds) 5647 in response to low-memory conditions. The range 5648 of permitted values is in the range 0:100000. 5649 5650 rcutree.rcu_divisor= [KNL] 5651 Set the shift-right count to use to compute 5652 the callback-invocation batch limit bl from 5653 the number of callbacks queued on this CPU. 5654 The result will be bounded below by the value of 5655 the rcutree.blimit kernel parameter. Every bl 5656 callbacks, the softirq handler will exit in 5657 order to allow the CPU to do other work. 5658 5659 Please note that this callback-invocation batch 5660 limit applies only to non-offloaded callback 5661 invocation. Offloaded callbacks are instead 5662 invoked in the context of an rcuoc kthread, which 5663 scheduler will preempt as it does any other task. 5664 5665 rcutree.rcu_fanout_exact= [KNL] 5666 Disable autobalancing of the rcu_node combining 5667 tree. This is used by rcutorture, and might 5668 possibly be useful for architectures having high 5669 cache-to-cache transfer latencies. 5670 5671 rcutree.rcu_fanout_leaf= [KNL] 5672 Change the number of CPUs assigned to each 5673 leaf rcu_node structure. Useful for very 5674 large systems, which will choose the value 64, 5675 and for NUMA systems with large remote-access 5676 latencies, which will choose a value aligned 5677 with the appropriate hardware boundaries. 5678 5679 rcutree.rcu_min_cached_objs= [KNL] 5680 Minimum number of objects which are cached and 5681 maintained per one CPU. Object size is equal 5682 to PAGE_SIZE. The cache allows to reduce the 5683 pressure to page allocator, also it makes the 5684 whole algorithm to behave better in low memory 5685 condition. 5686 5687 rcutree.rcu_nocb_gp_stride= [KNL] 5688 Set the number of NOCB callback kthreads in 5689 each group, which defaults to the square root 5690 of the number of CPUs. Larger numbers reduce 5691 the wakeup overhead on the global grace-period 5692 kthread, but increases that same overhead on 5693 each group's NOCB grace-period kthread. 5694 5695 rcutree.rcu_kick_kthreads= [KNL] 5696 Cause the grace-period kthread to get an extra 5697 wake_up() if it sleeps three times longer than 5698 it should at force-quiescent-state time. 5699 This wake_up() will be accompanied by a 5700 WARN_ONCE() splat and an ftrace_dump(). 5701 5702 rcutree.rcu_resched_ns= [KNL] 5703 Limit the time spend invoking a batch of RCU 5704 callbacks to the specified number of nanoseconds. 5705 By default, this limit is checked only once 5706 every 32 callbacks in order to limit the pain 5707 inflicted by local_clock() overhead. 5708 5709 rcutree.rcu_unlock_delay= [KNL] 5710 In CONFIG_RCU_STRICT_GRACE_PERIOD=y kernels, 5711 this specifies an rcu_read_unlock()-time delay 5712 in microseconds. This defaults to zero. 5713 Larger delays increase the probability of 5714 catching RCU pointer leaks, that is, buggy use 5715 of RCU-protected pointers after the relevant 5716 rcu_read_unlock() has completed. 5717 5718 rcutree.sysrq_rcu= [KNL] 5719 Commandeer a sysrq key to dump out Tree RCU's 5720 rcu_node tree with an eye towards determining 5721 why a new grace period has not yet started. 5722 5723 rcutree.use_softirq= [KNL] 5724 If set to zero, move all RCU_SOFTIRQ processing to 5725 per-CPU rcuc kthreads. Defaults to a non-zero 5726 value, meaning that RCU_SOFTIRQ is used by default. 5727 Specify rcutree.use_softirq=0 to use rcuc kthreads. 5728 5729 But note that CONFIG_PREEMPT_RT=y kernels disable 5730 this kernel boot parameter, forcibly setting it 5731 to zero. 5732 5733 rcutree.enable_rcu_lazy= [KNL] 5734 To save power, batch RCU callbacks and flush after 5735 delay, memory pressure or callback list growing too 5736 big. 5737 5738 rcutree.rcu_normal_wake_from_gp= [KNL] 5739 Reduces a latency of synchronize_rcu() call. This approach 5740 maintains its own track of synchronize_rcu() callers, so it 5741 does not interact with regular callbacks because it does not 5742 use a call_rcu[_hurry]() path. Please note, this is for a 5743 normal grace period. 5744 5745 How to enable it: 5746 5747 echo 1 > /sys/module/rcutree/parameters/rcu_normal_wake_from_gp 5748 or pass a boot parameter "rcutree.rcu_normal_wake_from_gp=1" 5749 5750 Default is 1 if num_possible_cpus() <= 16 and it is not explicitly 5751 disabled by the boot parameter passing 0. 5752 5753 rcuscale.gp_async= [KNL] 5754 Measure performance of asynchronous 5755 grace-period primitives such as call_rcu(). 5756 5757 rcuscale.gp_async_max= [KNL] 5758 Specify the maximum number of outstanding 5759 callbacks per writer thread. When a writer 5760 thread exceeds this limit, it invokes the 5761 corresponding flavor of rcu_barrier() to allow 5762 previously posted callbacks to drain. 5763 5764 rcuscale.gp_exp= [KNL] 5765 Measure performance of expedited synchronous 5766 grace-period primitives. 5767 5768 rcuscale.holdoff= [KNL] 5769 Set test-start holdoff period. The purpose of 5770 this parameter is to delay the start of the 5771 test until boot completes in order to avoid 5772 interference. 5773 5774 rcuscale.kfree_by_call_rcu= [KNL] 5775 In kernels built with CONFIG_RCU_LAZY=y, test 5776 call_rcu() instead of kfree_rcu(). 5777 5778 rcuscale.kfree_mult= [KNL] 5779 Instead of allocating an object of size kfree_obj, 5780 allocate one of kfree_mult * sizeof(kfree_obj). 5781 Defaults to 1. 5782 5783 rcuscale.kfree_rcu_test= [KNL] 5784 Set to measure performance of kfree_rcu() flooding. 5785 5786 rcuscale.kfree_rcu_test_double= [KNL] 5787 Test the double-argument variant of kfree_rcu(). 5788 If this parameter has the same value as 5789 rcuscale.kfree_rcu_test_single, both the single- 5790 and double-argument variants are tested. 5791 5792 rcuscale.kfree_rcu_test_single= [KNL] 5793 Test the single-argument variant of kfree_rcu(). 5794 If this parameter has the same value as 5795 rcuscale.kfree_rcu_test_double, both the single- 5796 and double-argument variants are tested. 5797 5798 rcuscale.kfree_nthreads= [KNL] 5799 The number of threads running loops of kfree_rcu(). 5800 5801 rcuscale.kfree_alloc_num= [KNL] 5802 Number of allocations and frees done in an iteration. 5803 5804 rcuscale.kfree_loops= [KNL] 5805 Number of loops doing rcuscale.kfree_alloc_num number 5806 of allocations and frees. 5807 5808 rcuscale.minruntime= [KNL] 5809 Set the minimum test run time in seconds. This 5810 does not affect the data-collection interval, 5811 but instead allows better measurement of things 5812 like CPU consumption. 5813 5814 rcuscale.nreaders= [KNL] 5815 Set number of RCU readers. The value -1 selects 5816 N, where N is the number of CPUs. A value 5817 "n" less than -1 selects N-n+1, where N is again 5818 the number of CPUs. For example, -2 selects N 5819 (the number of CPUs), -3 selects N+1, and so on. 5820 A value of "n" less than or equal to -N selects 5821 a single reader. 5822 5823 rcuscale.nwriters= [KNL] 5824 Set number of RCU writers. The values operate 5825 the same as for rcuscale.nreaders. 5826 N, where N is the number of CPUs 5827 5828 rcuscale.scale_type= [KNL] 5829 Specify the RCU implementation to test. 5830 5831 rcuscale.shutdown= [KNL] 5832 Shut the system down after performance tests 5833 complete. This is useful for hands-off automated 5834 testing. 5835 5836 rcuscale.verbose= [KNL] 5837 Enable additional printk() statements. 5838 5839 rcuscale.writer_holdoff= [KNL] 5840 Write-side holdoff between grace periods, 5841 in microseconds. The default of zero says 5842 no holdoff. 5843 5844 rcuscale.writer_holdoff_jiffies= [KNL] 5845 Additional write-side holdoff between grace 5846 periods, but in jiffies. The default of zero 5847 says no holdoff. 5848 5849 rcutorture.fqs_duration= [KNL] 5850 Set duration of force_quiescent_state bursts 5851 in microseconds. 5852 5853 rcutorture.fqs_holdoff= [KNL] 5854 Set holdoff time within force_quiescent_state bursts 5855 in microseconds. 5856 5857 rcutorture.fqs_stutter= [KNL] 5858 Set wait time between force_quiescent_state bursts 5859 in seconds. 5860 5861 rcutorture.fwd_progress= [KNL] 5862 Specifies the number of kthreads to be used 5863 for RCU grace-period forward-progress testing 5864 for the types of RCU supporting this notion. 5865 Defaults to 1 kthread, values less than zero or 5866 greater than the number of CPUs cause the number 5867 of CPUs to be used. 5868 5869 rcutorture.fwd_progress_div= [KNL] 5870 Specify the fraction of a CPU-stall-warning 5871 period to do tight-loop forward-progress testing. 5872 5873 rcutorture.fwd_progress_holdoff= [KNL] 5874 Number of seconds to wait between successive 5875 forward-progress tests. 5876 5877 rcutorture.fwd_progress_need_resched= [KNL] 5878 Enclose cond_resched() calls within checks for 5879 need_resched() during tight-loop forward-progress 5880 testing. 5881 5882 rcutorture.gp_cond= [KNL] 5883 Use conditional/asynchronous update-side 5884 normal-grace-period primitives, if available. 5885 5886 rcutorture.gp_cond_exp= [KNL] 5887 Use conditional/asynchronous update-side 5888 expedited-grace-period primitives, if available. 5889 5890 rcutorture.gp_cond_full= [KNL] 5891 Use conditional/asynchronous update-side 5892 normal-grace-period primitives that also take 5893 concurrent expedited grace periods into account, 5894 if available. 5895 5896 rcutorture.gp_cond_exp_full= [KNL] 5897 Use conditional/asynchronous update-side 5898 expedited-grace-period primitives that also take 5899 concurrent normal grace periods into account, 5900 if available. 5901 5902 rcutorture.gp_cond_wi= [KNL] 5903 Nominal wait interval for normal conditional 5904 grace periods (specified by rcutorture's 5905 gp_cond and gp_cond_full module parameters), 5906 in microseconds. The actual wait interval will 5907 be randomly selected to nanosecond granularity up 5908 to this wait interval. Defaults to 16 jiffies, 5909 for example, 16,000 microseconds on a system 5910 with HZ=1000. 5911 5912 rcutorture.gp_cond_wi_exp= [KNL] 5913 Nominal wait interval for expedited conditional 5914 grace periods (specified by rcutorture's 5915 gp_cond_exp and gp_cond_exp_full module 5916 parameters), in microseconds. The actual wait 5917 interval will be randomly selected to nanosecond 5918 granularity up to this wait interval. Defaults to 5919 128 microseconds. 5920 5921 rcutorture.gp_exp= [KNL] 5922 Use expedited update-side primitives, if available. 5923 5924 rcutorture.gp_normal= [KNL] 5925 Use normal (non-expedited) asynchronous 5926 update-side primitives, if available. 5927 5928 rcutorture.gp_poll= [KNL] 5929 Use polled update-side normal-grace-period 5930 primitives, if available. 5931 5932 rcutorture.gp_poll_exp= [KNL] 5933 Use polled update-side expedited-grace-period 5934 primitives, if available. 5935 5936 rcutorture.gp_poll_full= [KNL] 5937 Use polled update-side normal-grace-period 5938 primitives that also take concurrent expedited 5939 grace periods into account, if available. 5940 5941 rcutorture.gp_poll_exp_full= [KNL] 5942 Use polled update-side expedited-grace-period 5943 primitives that also take concurrent normal 5944 grace periods into account, if available. 5945 5946 rcutorture.gp_poll_wi= [KNL] 5947 Nominal wait interval for normal conditional 5948 grace periods (specified by rcutorture's 5949 gp_poll and gp_poll_full module parameters), 5950 in microseconds. The actual wait interval will 5951 be randomly selected to nanosecond granularity up 5952 to this wait interval. Defaults to 16 jiffies, 5953 for example, 16,000 microseconds on a system 5954 with HZ=1000. 5955 5956 rcutorture.gp_poll_wi_exp= [KNL] 5957 Nominal wait interval for expedited conditional 5958 grace periods (specified by rcutorture's 5959 gp_poll_exp and gp_poll_exp_full module 5960 parameters), in microseconds. The actual wait 5961 interval will be randomly selected to nanosecond 5962 granularity up to this wait interval. Defaults to 5963 128 microseconds. 5964 5965 rcutorture.gp_sync= [KNL] 5966 Use normal (non-expedited) synchronous 5967 update-side primitives, if available. If all 5968 of rcutorture.gp_cond=, rcutorture.gp_exp=, 5969 rcutorture.gp_normal=, and rcutorture.gp_sync= 5970 are zero, rcutorture acts as if is interpreted 5971 they are all non-zero. 5972 5973 rcutorture.gpwrap_lag= [KNL] 5974 Enable grace-period wrap lag testing. Setting 5975 to false prevents the gpwrap lag test from 5976 running. Default is true. 5977 5978 rcutorture.gpwrap_lag_gps= [KNL] 5979 Set the value for grace-period wrap lag during 5980 active lag testing periods. This controls how many 5981 grace periods differences we tolerate between 5982 rdp and rnp's gp_seq before setting overflow flag. 5983 The default is always set to 8. 5984 5985 rcutorture.gpwrap_lag_cycle_mins= [KNL] 5986 Set the total cycle duration for gpwrap lag 5987 testing in minutes. This is the total time for 5988 one complete cycle of active and inactive 5989 testing periods. Default is 30 minutes. 5990 5991 rcutorture.gpwrap_lag_active_mins= [KNL] 5992 Set the duration for which gpwrap lag is active 5993 within each cycle, in minutes. During this time, 5994 the grace-period wrap lag will be set to the 5995 value specified by gpwrap_lag_gps. Default is 5996 5 minutes. 5997 5998 rcutorture.irqreader= [KNL] 5999 Run RCU readers from irq handlers, or, more 6000 accurately, from a timer handler. Not all RCU 6001 flavors take kindly to this sort of thing. 6002 6003 rcutorture.leakpointer= [KNL] 6004 Leak an RCU-protected pointer out of the reader. 6005 This can of course result in splats, and is 6006 intended to test the ability of things like 6007 CONFIG_RCU_STRICT_GRACE_PERIOD=y to detect 6008 such leaks. 6009 6010 rcutorture.n_barrier_cbs= [KNL] 6011 Set callbacks/threads for rcu_barrier() testing. 6012 6013 rcutorture.nfakewriters= [KNL] 6014 Set number of concurrent RCU writers. These just 6015 stress RCU, they don't participate in the actual 6016 test, hence the "fake". 6017 6018 rcutorture.nocbs_nthreads= [KNL] 6019 Set number of RCU callback-offload togglers. 6020 Zero (the default) disables toggling. 6021 6022 rcutorture.nocbs_toggle= [KNL] 6023 Set the delay in milliseconds between successive 6024 callback-offload toggling attempts. 6025 6026 rcutorture.nreaders= [KNL] 6027 Set number of RCU readers. The value -1 selects 6028 N-1, where N is the number of CPUs. A value 6029 "n" less than -1 selects N-n-2, where N is again 6030 the number of CPUs. For example, -2 selects N 6031 (the number of CPUs), -3 selects N+1, and so on. 6032 6033 rcutorture.object_debug= [KNL] 6034 Enable debug-object double-call_rcu() testing. 6035 6036 rcutorture.onoff_holdoff= [KNL] 6037 Set time (s) after boot for CPU-hotplug testing. 6038 6039 rcutorture.onoff_interval= [KNL] 6040 Set time (jiffies) between CPU-hotplug operations, 6041 or zero to disable CPU-hotplug testing. 6042 6043 rcutorture.preempt_duration= [KNL] 6044 Set duration (in milliseconds) of preemptions 6045 by a high-priority FIFO real-time task. Set to 6046 zero (the default) to disable. The CPUs to 6047 preempt are selected randomly from the set that 6048 are online at a given point in time. Races with 6049 CPUs going offline are ignored, with that attempt 6050 at preemption skipped. 6051 6052 rcutorture.preempt_interval= [KNL] 6053 Set interval (in milliseconds, defaulting to one 6054 second) between preemptions by a high-priority 6055 FIFO real-time task. This delay is mediated 6056 by an hrtimer and is further fuzzed to avoid 6057 inadvertent synchronizations. 6058 6059 rcutorture.read_exit_burst= [KNL] 6060 The number of times in a given read-then-exit 6061 episode that a set of read-then-exit kthreads 6062 is spawned. 6063 6064 rcutorture.read_exit_delay= [KNL] 6065 The delay, in seconds, between successive 6066 read-then-exit testing episodes. 6067 6068 rcutorture.reader_flavor= [KNL] 6069 A bit mask indicating which readers to use. 6070 If there is more than one bit set, the readers 6071 are entered from low-order bit up, and are 6072 exited in the opposite order. For SRCU, the 6073 0x1 bit is normal readers, 0x2 NMI-safe readers, 6074 and 0x4 light-weight readers. 6075 6076 rcutorture.shuffle_interval= [KNL] 6077 Set task-shuffle interval (s). Shuffling tasks 6078 allows some CPUs to go into dyntick-idle mode 6079 during the rcutorture test. 6080 6081 rcutorture.shutdown_secs= [KNL] 6082 Set time (s) after boot system shutdown. This 6083 is useful for hands-off automated testing. 6084 6085 rcutorture.stall_cpu= [KNL] 6086 Duration of CPU stall (s) to test RCU CPU stall 6087 warnings, zero to disable. 6088 6089 rcutorture.stall_cpu_block= [KNL] 6090 Sleep while stalling if set. This will result 6091 in warnings from preemptible RCU in addition to 6092 any other stall-related activity. Note that 6093 in kernels built with CONFIG_PREEMPTION=n and 6094 CONFIG_PREEMPT_COUNT=y, this parameter will 6095 cause the CPU to pass through a quiescent state. 6096 Given CONFIG_PREEMPTION=n, this will suppress 6097 RCU CPU stall warnings, but will instead result 6098 in scheduling-while-atomic splats. 6099 6100 Use of this module parameter results in splats. 6101 6102 6103 rcutorture.stall_cpu_holdoff= [KNL] 6104 Time to wait (s) after boot before inducing stall. 6105 6106 rcutorture.stall_cpu_irqsoff= [KNL] 6107 Disable interrupts while stalling if set, but only 6108 on the first stall in the set. 6109 6110 rcutorture.stall_cpu_repeat= [KNL] 6111 Number of times to repeat the stall sequence, 6112 so that rcutorture.stall_cpu_repeat=3 will result 6113 in four stall sequences. 6114 6115 rcutorture.stall_gp_kthread= [KNL] 6116 Duration (s) of forced sleep within RCU 6117 grace-period kthread to test RCU CPU stall 6118 warnings, zero to disable. If both stall_cpu 6119 and stall_gp_kthread are specified, the 6120 kthread is starved first, then the CPU. 6121 6122 rcutorture.stat_interval= [KNL] 6123 Time (s) between statistics printk()s. 6124 6125 rcutorture.stutter= [KNL] 6126 Time (s) to stutter testing, for example, specifying 6127 five seconds causes the test to run for five seconds, 6128 wait for five seconds, and so on. This tests RCU's 6129 ability to transition abruptly to and from idle. 6130 6131 rcutorture.test_boost= [KNL] 6132 Test RCU priority boosting? 0=no, 1=maybe, 2=yes. 6133 "Maybe" means test if the RCU implementation 6134 under test support RCU priority boosting. 6135 6136 rcutorture.test_boost_duration= [KNL] 6137 Duration (s) of each individual boost test. 6138 6139 rcutorture.test_boost_holdoff= [KNL] 6140 Holdoff time (s) from start of test to the start 6141 of RCU priority-boost testing. Defaults to zero, 6142 that is, no holdoff. 6143 6144 rcutorture.test_boost_interval= [KNL] 6145 Interval (s) between each boost test. 6146 6147 rcutorture.test_no_idle_hz= [KNL] 6148 Test RCU's dyntick-idle handling. See also the 6149 rcutorture.shuffle_interval parameter. 6150 6151 rcutorture.torture_type= [KNL] 6152 Specify the RCU implementation to test. 6153 6154 rcutorture.verbose= [KNL] 6155 Enable additional printk() statements. 6156 6157 rcupdate.rcu_cpu_stall_ftrace_dump= [KNL] 6158 Dump ftrace buffer after reporting RCU CPU 6159 stall warning. 6160 6161 rcupdate.rcu_cpu_stall_notifiers= [KNL] 6162 Provide RCU CPU stall notifiers, but see the 6163 warnings in the RCU_CPU_STALL_NOTIFIER Kconfig 6164 option's help text. TL;DR: You almost certainly 6165 do not want rcupdate.rcu_cpu_stall_notifiers. 6166 6167 rcupdate.rcu_cpu_stall_suppress= [KNL] 6168 Suppress RCU CPU stall warning messages. 6169 6170 rcupdate.rcu_cpu_stall_suppress_at_boot= [KNL] 6171 Suppress RCU CPU stall warning messages and 6172 rcutorture writer stall warnings that occur 6173 during early boot, that is, during the time 6174 before the init task is spawned. 6175 6176 rcupdate.rcu_cpu_stall_timeout= [KNL] 6177 Set timeout for RCU CPU stall warning messages. 6178 The value is in seconds and the maximum allowed 6179 value is 300 seconds. 6180 6181 rcupdate.rcu_exp_cpu_stall_timeout= [KNL] 6182 Set timeout for expedited RCU CPU stall warning 6183 messages. The value is in milliseconds 6184 and the maximum allowed value is 21000 6185 milliseconds. Please note that this value is 6186 adjusted to an arch timer tick resolution. 6187 Setting this to zero causes the value from 6188 rcupdate.rcu_cpu_stall_timeout to be used (after 6189 conversion from seconds to milliseconds). 6190 6191 rcupdate.rcu_cpu_stall_cputime= [KNL] 6192 Provide statistics on the cputime and count of 6193 interrupts and tasks during the sampling period. For 6194 multiple continuous RCU stalls, all sampling periods 6195 begin at half of the first RCU stall timeout. 6196 6197 rcupdate.rcu_exp_stall_task_details= [KNL] 6198 Print stack dumps of any tasks blocking the 6199 current expedited RCU grace period during an 6200 expedited RCU CPU stall warning. 6201 6202 rcupdate.rcu_expedited= [KNL] 6203 Use expedited grace-period primitives, for 6204 example, synchronize_rcu_expedited() instead 6205 of synchronize_rcu(). This reduces latency, 6206 but can increase CPU utilization, degrade 6207 real-time latency, and degrade energy efficiency. 6208 No effect on CONFIG_TINY_RCU kernels. 6209 6210 rcupdate.rcu_normal= [KNL] 6211 Use only normal grace-period primitives, 6212 for example, synchronize_rcu() instead of 6213 synchronize_rcu_expedited(). This improves 6214 real-time latency, CPU utilization, and 6215 energy efficiency, but can expose users to 6216 increased grace-period latency. This parameter 6217 overrides rcupdate.rcu_expedited. No effect on 6218 CONFIG_TINY_RCU kernels. 6219 6220 rcupdate.rcu_normal_after_boot= [KNL] 6221 Once boot has completed (that is, after 6222 rcu_end_inkernel_boot() has been invoked), use 6223 only normal grace-period primitives. No effect 6224 on CONFIG_TINY_RCU kernels. 6225 6226 But note that CONFIG_PREEMPT_RT=y kernels enables 6227 this kernel boot parameter, forcibly setting 6228 it to the value one, that is, converting any 6229 post-boot attempt at an expedited RCU grace 6230 period to instead use normal non-expedited 6231 grace-period processing. 6232 6233 rcupdate.rcu_task_collapse_lim= [KNL] 6234 Set the maximum number of callbacks present 6235 at the beginning of a grace period that allows 6236 the RCU Tasks flavors to collapse back to using 6237 a single callback queue. This switching only 6238 occurs when rcupdate.rcu_task_enqueue_lim is 6239 set to the default value of -1. 6240 6241 rcupdate.rcu_task_contend_lim= [KNL] 6242 Set the minimum number of callback-queuing-time 6243 lock-contention events per jiffy required to 6244 cause the RCU Tasks flavors to switch to per-CPU 6245 callback queuing. This switching only occurs 6246 when rcupdate.rcu_task_enqueue_lim is set to 6247 the default value of -1. 6248 6249 rcupdate.rcu_task_enqueue_lim= [KNL] 6250 Set the number of callback queues to use for the 6251 RCU Tasks family of RCU flavors. The default 6252 of -1 allows this to be automatically (and 6253 dynamically) adjusted. This parameter is intended 6254 for use in testing. 6255 6256 rcupdate.rcu_task_ipi_delay= [KNL] 6257 Set time in jiffies during which RCU tasks will 6258 avoid sending IPIs, starting with the beginning 6259 of a given grace period. Setting a large 6260 number avoids disturbing real-time workloads, 6261 but lengthens grace periods. 6262 6263 rcupdate.rcu_task_lazy_lim= [KNL] 6264 Number of callbacks on a given CPU that will 6265 cancel laziness on that CPU. Use -1 to disable 6266 cancellation of laziness, but be advised that 6267 doing so increases the danger of OOM due to 6268 callback flooding. 6269 6270 rcupdate.rcu_task_stall_info= [KNL] 6271 Set initial timeout in jiffies for RCU task stall 6272 informational messages, which give some indication 6273 of the problem for those not patient enough to 6274 wait for ten minutes. Informational messages are 6275 only printed prior to the stall-warning message 6276 for a given grace period. Disable with a value 6277 less than or equal to zero. Defaults to ten 6278 seconds. A change in value does not take effect 6279 until the beginning of the next grace period. 6280 6281 rcupdate.rcu_task_stall_info_mult= [KNL] 6282 Multiplier for time interval between successive 6283 RCU task stall informational messages for a given 6284 RCU tasks grace period. This value is clamped 6285 to one through ten, inclusive. It defaults to 6286 the value three, so that the first informational 6287 message is printed 10 seconds into the grace 6288 period, the second at 40 seconds, the third at 6289 160 seconds, and then the stall warning at 600 6290 seconds would prevent a fourth at 640 seconds. 6291 6292 rcupdate.rcu_task_stall_timeout= [KNL] 6293 Set timeout in jiffies for RCU task stall 6294 warning messages. Disable with a value less 6295 than or equal to zero. Defaults to ten minutes. 6296 A change in value does not take effect until 6297 the beginning of the next grace period. 6298 6299 rcupdate.rcu_tasks_lazy_ms= [KNL] 6300 Set timeout in milliseconds RCU Tasks asynchronous 6301 callback batching for call_rcu_tasks(). 6302 A negative value will take the default. A value 6303 of zero will disable batching. Batching is 6304 always disabled for synchronize_rcu_tasks(). 6305 6306 rcupdate.rcu_tasks_trace_lazy_ms= [KNL] 6307 Set timeout in milliseconds RCU Tasks 6308 Trace asynchronous callback batching for 6309 call_rcu_tasks_trace(). A negative value 6310 will take the default. A value of zero will 6311 disable batching. Batching is always disabled 6312 for synchronize_rcu_tasks_trace(). 6313 6314 rcupdate.rcu_self_test= [KNL] 6315 Run the RCU early boot self tests 6316 6317 rdinit= [KNL] 6318 Format: <full_path> 6319 Run specified binary instead of /init from the ramdisk, 6320 used for early userspace startup. See initrd. 6321 6322 rdrand= [X86,EARLY] 6323 force - Override the decision by the kernel to hide the 6324 advertisement of RDRAND support (this affects 6325 certain AMD processors because of buggy BIOS 6326 support, specifically around the suspend/resume 6327 path). 6328 6329 rdt= [HW,X86,RDT] 6330 Turn on/off individual RDT features. List is: 6331 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp, 6332 mba, smba, bmec, abmc, sdciae. 6333 E.g. to turn on cmt and turn off mba use: 6334 rdt=cmt,!mba 6335 6336 reboot= [KNL] 6337 Format (x86 or x86_64): 6338 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] | d[efault] \ 6339 [[,]s[mp]#### \ 6340 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \ 6341 [[,]f[orce] 6342 Where reboot_mode is one of warm (soft) or cold (hard) or gpio 6343 (prefix with 'panic_' to set mode for panic 6344 reboot only), 6345 reboot_type is one of bios, acpi, kbd, triple, efi, or pci, 6346 reboot_force is either force or not specified, 6347 reboot_cpu is s[mp]#### with #### being the processor 6348 to be used for rebooting. 6349 6350 acpi 6351 Use the ACPI RESET_REG in the FADT. If ACPI is not 6352 configured or the ACPI reset does not work, the reboot 6353 path attempts the reset using the keyboard controller. 6354 6355 bios 6356 Use the CPU reboot vector for warm reset 6357 6358 cold 6359 Set the cold reboot flag 6360 6361 default 6362 There are some built-in platform specific "quirks" 6363 - you may see: "reboot: <name> series board detected. 6364 Selecting <type> for reboots." In the case where you 6365 think the quirk is in error (e.g. you have newer BIOS, 6366 or newer board) using this option will ignore the 6367 built-in quirk table, and use the generic default 6368 reboot actions. 6369 6370 efi 6371 Use efi reset_system runtime service. If EFI is not 6372 configured or the EFI reset does not work, the reboot 6373 path attempts the reset using the keyboard controller. 6374 6375 force 6376 Don't stop other CPUs on reboot. This can make reboot 6377 more reliable in some cases. 6378 6379 kbd 6380 Use the keyboard controller. cold reset (default) 6381 6382 pci 6383 Use a write to the PCI config space register 0xcf9 to 6384 trigger reboot. 6385 6386 triple 6387 Force a triple fault (init) 6388 6389 warm 6390 Don't set the cold reboot flag 6391 6392 Using warm reset will be much faster especially on big 6393 memory systems because the BIOS will not go through 6394 the memory check. Disadvantage is that not all 6395 hardware will be completely reinitialized on reboot so 6396 there may be boot problems on some systems. 6397 6398 6399 refscale.holdoff= [KNL] 6400 Set test-start holdoff period. The purpose of 6401 this parameter is to delay the start of the 6402 test until boot completes in order to avoid 6403 interference. 6404 6405 refscale.lookup_instances= [KNL] 6406 Number of data elements to use for the forms of 6407 SLAB_TYPESAFE_BY_RCU testing. A negative number 6408 is negated and multiplied by nr_cpu_ids, while 6409 zero specifies nr_cpu_ids. 6410 6411 refscale.loops= [KNL] 6412 Set the number of loops over the synchronization 6413 primitive under test. Increasing this number 6414 reduces noise due to loop start/end overhead, 6415 but the default has already reduced the per-pass 6416 noise to a handful of picoseconds on ca. 2020 6417 x86 laptops. 6418 6419 refscale.nreaders= [KNL] 6420 Set number of readers. The default value of -1 6421 selects N, where N is roughly 75% of the number 6422 of CPUs. A value of zero is an interesting choice. 6423 6424 refscale.nruns= [KNL] 6425 Set number of runs, each of which is dumped onto 6426 the console log. 6427 6428 refscale.readdelay= [KNL] 6429 Set the read-side critical-section duration, 6430 measured in microseconds. 6431 6432 refscale.scale_type= [KNL] 6433 Specify the read-protection implementation to test. 6434 6435 refscale.shutdown= [KNL] 6436 Shut down the system at the end of the performance 6437 test. This defaults to 1 (shut it down) when 6438 refscale is built into the kernel and to 0 (leave 6439 it running) when refscale is built as a module. 6440 6441 refscale.verbose= [KNL] 6442 Enable additional printk() statements. 6443 6444 refscale.verbose_batched= [KNL] 6445 Batch the additional printk() statements. If zero 6446 (the default) or negative, print everything. Otherwise, 6447 print every Nth verbose statement, where N is the value 6448 specified. 6449 6450 regulator_ignore_unused 6451 [REGULATOR] 6452 Prevents regulator framework from disabling regulators 6453 that are unused, due no driver claiming them. This may 6454 be useful for debug and development, but should not be 6455 needed on a platform with proper driver support. 6456 6457 relax_domain_level= 6458 [KNL, SMP] Set scheduler's default relax_domain_level. 6459 See Documentation/admin-guide/cgroup-v1/cpusets.rst. 6460 6461 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory 6462 Format: <base1>,<size1>[,<base2>,<size2>,...] 6463 Reserve I/O ports or memory so the kernel won't use 6464 them. If <base> is less than 0x10000, the region 6465 is assumed to be I/O ports; otherwise it is memory. 6466 6467 reserve_mem= [RAM] 6468 Format: nn[KMG]:<align>:<label> 6469 Reserve physical memory and label it with a name that 6470 other subsystems can use to access it. This is typically 6471 used for systems that do not wipe the RAM, and this command 6472 line will try to reserve the same physical memory on 6473 soft reboots. Note, it is not guaranteed to be the same 6474 location. For example, if anything about the system changes 6475 or if booting a different kernel. It can also fail if KASLR 6476 places the kernel at the location of where the RAM reservation 6477 was from a previous boot, the new reservation will be at a 6478 different location. 6479 Any subsystem using this feature must add a way to verify 6480 that the contents of the physical memory is from a previous 6481 boot, as there may be cases where the memory will not be 6482 located at the same location. 6483 6484 The format is size:align:label for example, to request 6485 12 megabytes of 4096 alignment for ramoops: 6486 6487 reserve_mem=12M:4096:oops ramoops.mem_name=oops 6488 6489 reservetop= [X86-32,EARLY] 6490 Format: nn[KMG] 6491 Reserves a hole at the top of the kernel virtual 6492 address space. 6493 6494 reset_devices [KNL] Force drivers to reset the underlying device 6495 during initialization. 6496 6497 resume= [SWSUSP] 6498 Specify the partition device for software suspend 6499 Format: 6500 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>} 6501 6502 resume_offset= [SWSUSP] 6503 Specify the offset from the beginning of the partition 6504 given by "resume=" at which the swap header is located, 6505 in <PAGE_SIZE> units (needed only for swap files). 6506 See Documentation/power/swsusp-and-swap-files.rst 6507 6508 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to 6509 read the resume files 6510 6511 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up. 6512 Useful for devices that are detected asynchronously 6513 (e.g. USB and MMC devices). 6514 6515 retain_initrd [RAM] Keep initrd memory after extraction. After boot, it will 6516 be accessible via /sys/firmware/initrd. 6517 6518 retbleed= [X86] Control mitigation of RETBleed (Arbitrary 6519 Speculative Code Execution with Return Instructions) 6520 vulnerability. 6521 6522 AMD-based UNRET and IBPB mitigations alone do not stop 6523 sibling threads from influencing the predictions of other 6524 sibling threads. For that reason, STIBP is used on pro- 6525 cessors that support it, and mitigate SMT on processors 6526 that don't. 6527 6528 off - no mitigation 6529 auto - automatically select a mitigation 6530 auto,nosmt - automatically select a mitigation, 6531 disabling SMT if necessary for 6532 the full mitigation (only on Zen1 6533 and older without STIBP). 6534 ibpb - On AMD, mitigate short speculation 6535 windows on basic block boundaries too. 6536 Safe, highest perf impact. It also 6537 enables STIBP if present. Not suitable 6538 on Intel. 6539 ibpb,nosmt - Like "ibpb" above but will disable SMT 6540 when STIBP is not available. This is 6541 the alternative for systems which do not 6542 have STIBP. 6543 unret - Force enable untrained return thunks, 6544 only effective on AMD f15h-f17h based 6545 systems. 6546 unret,nosmt - Like unret, but will disable SMT when STIBP 6547 is not available. This is the alternative for 6548 systems which do not have STIBP. 6549 6550 Selecting 'auto' will choose a mitigation method at run 6551 time according to the CPU. 6552 6553 Not specifying this option is equivalent to retbleed=auto. 6554 6555 rfkill.default_state= 6556 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm, 6557 etc. communication is blocked by default. 6558 1 Unblocked. 6559 6560 rfkill.master_switch_mode= 6561 0 The "airplane mode" button does nothing. 6562 1 The "airplane mode" button toggles between everything 6563 blocked and the previous configuration. 6564 2 The "airplane mode" button toggles between everything 6565 blocked and everything unblocked. 6566 6567 ring3mwait=disable 6568 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported 6569 CPUs. 6570 6571 riscv_isa_fallback [RISCV,EARLY] 6572 When CONFIG_RISCV_ISA_FALLBACK is not enabled, permit 6573 falling back to detecting extension support by parsing 6574 "riscv,isa" property on devicetree systems when the 6575 replacement properties are not found. See the Kconfig 6576 entry for RISCV_ISA_FALLBACK. 6577 6578 ro [KNL] Mount root device read-only on boot 6579 6580 rodata= [KNL,EARLY] 6581 on Mark read-only kernel memory as read-only (default). 6582 off Leave read-only kernel memory writable for debugging. 6583 noalias Mark read-only kernel memory as read-only but retain 6584 writable aliases in the direct map for regions outside 6585 of the kernel image. [arm64] 6586 6587 rockchip.usb_uart 6588 [EARLY] 6589 Enable the uart passthrough on the designated usb port 6590 on Rockchip SoCs. When active, the signals of the 6591 debug-uart get routed to the D+ and D- pins of the usb 6592 port and the regular usb controller gets disabled. 6593 6594 root= [KNL] Root filesystem 6595 Usually this is a block device specifier of some kind, 6596 see the early_lookup_bdev comment in 6597 block/early-lookup.c for details. 6598 Alternatively this can be "ram" for the legacy initial 6599 ramdisk, "nfs" and "cifs" for root on a network file 6600 system, or "mtd" and "ubi" for mounting from raw flash. 6601 6602 rootdelay= [KNL] Delay (in seconds) to pause before attempting to 6603 mount the root filesystem 6604 6605 rootflags= [KNL] Set root filesystem mount option string 6606 6607 initramfs_options= [KNL] 6608 Specify mount options for for the initramfs mount. 6609 6610 rootfstype= [KNL] Set root filesystem type 6611 6612 rootwait [KNL] Wait (indefinitely) for root device to show up. 6613 Useful for devices that are detected asynchronously 6614 (e.g. USB and MMC devices). 6615 6616 rootwait= [KNL] Maximum time (in seconds) to wait for root device 6617 to show up before attempting to mount the root 6618 filesystem. 6619 6620 rproc_mem=nn[KMG][@address] 6621 [KNL,ARM,CMA] Remoteproc physical memory block. 6622 Memory area to be used by remote processor image, 6623 managed by CMA. 6624 6625 rseq_debug= [KNL] Enable or disable restartable sequence 6626 debug mode. Defaults to CONFIG_RSEQ_DEBUG_DEFAULT_ENABLE. 6627 Format: <bool> 6628 6629 rt_group_sched= [KNL] Enable or disable SCHED_RR/FIFO group scheduling 6630 when CONFIG_RT_GROUP_SCHED=y. Defaults to 6631 !CONFIG_RT_GROUP_SCHED_DEFAULT_DISABLED. 6632 Format: <bool> 6633 6634 rw [KNL] Mount root device read-write on boot 6635 6636 S [KNL] Run init in single mode 6637 6638 s390_iommu= [HW,S390] 6639 Set s390 IOTLB flushing mode 6640 strict 6641 With strict flushing every unmap operation will result 6642 in an IOTLB flush. Default is lazy flushing before 6643 reuse, which is faster. Deprecated, equivalent to 6644 iommu.strict=1. 6645 6646 s390_iommu_aperture= [KNL,S390] 6647 Specifies the size of the per device DMA address space 6648 accessible through the DMA and IOMMU APIs as a decimal 6649 factor of the size of main memory. 6650 The default is 1 meaning that one can concurrently use 6651 as many DMA addresses as physical memory is installed, 6652 if supported by hardware, and thus map all of memory 6653 once. With a value of 2 one can map all of memory twice 6654 and so on. As a special case a factor of 0 imposes no 6655 restrictions other than those given by hardware at the 6656 cost of significant additional memory use for tables. 6657 6658 sa1100ir [NET] 6659 See drivers/net/irda/sa1100_ir.c. 6660 6661 sched_proxy_exec= [KNL] 6662 Enables or disables "proxy execution" style 6663 solution to mutex-based priority inversion. 6664 Format: <bool> 6665 6666 sched_verbose [KNL,EARLY] Enables verbose scheduler debug messages. 6667 6668 schedstats= [KNL,X86] Enable or disable scheduled statistics. 6669 Allowed values are enable and disable. This feature 6670 incurs a small amount of overhead in the scheduler 6671 but is useful for debugging and performance tuning. 6672 6673 sched_thermal_decay_shift= 6674 [Deprecated] 6675 [KNL, SMP] Set a decay shift for scheduler thermal 6676 pressure signal. Thermal pressure signal follows the 6677 default decay period of other scheduler pelt 6678 signals(usually 32 ms but configurable). Setting 6679 sched_thermal_decay_shift will left shift the decay 6680 period for the thermal pressure signal by the shift 6681 value. 6682 i.e. with the default pelt decay period of 32 ms 6683 sched_thermal_decay_shift thermal pressure decay pr 6684 1 64 ms 6685 2 128 ms 6686 and so on. 6687 Format: integer between 0 and 10 6688 Default is 0. 6689 6690 scftorture.holdoff= [KNL] 6691 Number of seconds to hold off before starting 6692 test. Defaults to zero for module insertion and 6693 to 10 seconds for built-in smp_call_function() 6694 tests. 6695 6696 scftorture.longwait= [KNL] 6697 Request ridiculously long waits randomly selected 6698 up to the chosen limit in seconds. Zero (the 6699 default) disables this feature. Please note 6700 that requesting even small non-zero numbers of 6701 seconds can result in RCU CPU stall warnings, 6702 softlockup complaints, and so on. 6703 6704 scftorture.nthreads= [KNL] 6705 Number of kthreads to spawn to invoke the 6706 smp_call_function() family of functions. 6707 The default of -1 specifies a number of kthreads 6708 equal to the number of CPUs. 6709 6710 scftorture.onoff_holdoff= [KNL] 6711 Number seconds to wait after the start of the 6712 test before initiating CPU-hotplug operations. 6713 6714 scftorture.onoff_interval= [KNL] 6715 Number seconds to wait between successive 6716 CPU-hotplug operations. Specifying zero (which 6717 is the default) disables CPU-hotplug operations. 6718 6719 scftorture.shutdown_secs= [KNL] 6720 The number of seconds following the start of the 6721 test after which to shut down the system. The 6722 default of zero avoids shutting down the system. 6723 Non-zero values are useful for automated tests. 6724 6725 scftorture.stat_interval= [KNL] 6726 The number of seconds between outputting the 6727 current test statistics to the console. A value 6728 of zero disables statistics output. 6729 6730 scftorture.stutter_cpus= [KNL] 6731 The number of jiffies to wait between each change 6732 to the set of CPUs under test. 6733 6734 scftorture.use_cpus_read_lock= [KNL] 6735 Use use_cpus_read_lock() instead of the default 6736 preempt_disable() to disable CPU hotplug 6737 while invoking one of the smp_call_function*() 6738 functions. 6739 6740 scftorture.verbose= [KNL] 6741 Enable additional printk() statements. 6742 6743 scftorture.weight_single= [KNL] 6744 The probability weighting to use for the 6745 smp_call_function_single() function with a zero 6746 "wait" parameter. A value of -1 selects the 6747 default if all other weights are -1. However, 6748 if at least one weight has some other value, a 6749 value of -1 will instead select a weight of zero. 6750 6751 scftorture.weight_single_wait= [KNL] 6752 The probability weighting to use for the 6753 smp_call_function_single() function with a 6754 non-zero "wait" parameter. See weight_single. 6755 6756 scftorture.weight_many= [KNL] 6757 The probability weighting to use for the 6758 smp_call_function_many() function with a zero 6759 "wait" parameter. See weight_single. 6760 Note well that setting a high probability for 6761 this weighting can place serious IPI load 6762 on the system. 6763 6764 scftorture.weight_many_wait= [KNL] 6765 The probability weighting to use for the 6766 smp_call_function_many() function with a 6767 non-zero "wait" parameter. See weight_single 6768 and weight_many. 6769 6770 scftorture.weight_all= [KNL] 6771 The probability weighting to use for the 6772 smp_call_function_all() function with a zero 6773 "wait" parameter. See weight_single and 6774 weight_many. 6775 6776 scftorture.weight_all_wait= [KNL] 6777 The probability weighting to use for the 6778 smp_call_function_all() function with a 6779 non-zero "wait" parameter. See weight_single 6780 and weight_many. 6781 6782 sdw_mclk_divider=[SDW] 6783 Specify the MCLK divider for Intel SoundWire buses in 6784 case the BIOS does not provide the clock rate properly. 6785 6786 skew_tick= [KNL,EARLY] Offset the periodic timer tick per cpu to mitigate 6787 xtime_lock contention on larger systems, and/or RCU lock 6788 contention on all systems with CONFIG_MAXSMP set. 6789 Format: { "0" | "1" } 6790 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1" 6791 1 -- enable. 6792 Note: increases power consumption, thus should only be 6793 enabled if running jitter sensitive (HPC/RT) workloads. 6794 6795 security= [SECURITY] Choose a legacy "major" security module to 6796 enable at boot. This has been deprecated by the 6797 "lsm=" parameter. 6798 6799 selinux= [SELINUX] Disable or enable SELinux at boot time. 6800 Format: { "0" | "1" } 6801 See security/selinux/Kconfig help text. 6802 0 -- disable. 6803 1 -- enable. 6804 Default value is 1. 6805 6806 serialnumber [BUGS=X86-32] 6807 6808 sev=option[,option...] [X86-64] 6809 6810 debug 6811 Enable debug messages. 6812 6813 nosnp 6814 Do not enable SEV-SNP (applies to host/hypervisor 6815 only). Setting 'nosnp' avoids the RMP check overhead 6816 in memory accesses when users do not want to run 6817 SEV-SNP guests. 6818 6819 shapers= [NET] 6820 Maximal number of shapers. 6821 6822 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller 6823 Limit apic dumping. The parameter defines the maximal 6824 number of local apics being dumped. Also it is possible 6825 to set it to "all" by meaning -- no limit here. 6826 Format: { 1 (default) | 2 | ... | all }. 6827 The parameter valid if only apic=debug or 6828 apic=verbose is specified. 6829 Example: apic=debug show_lapic=all 6830 6831 slab_debug[=options[,slabs][;[options[,slabs]]...] [MM] 6832 Enabling slab_debug allows one to determine the 6833 culprit if slab objects become corrupted. Enabling 6834 slab_debug can create guard zones around objects and 6835 may poison objects when not in use. Also tracks the 6836 last alloc / free. For more information see 6837 Documentation/admin-guide/mm/slab.rst. 6838 (slub_debug legacy name also accepted for now) 6839 6840 Using this option implies the "no_hash_pointers" 6841 option which can be undone by adding the 6842 "hash_pointers=always" option. 6843 6844 slab_max_order= [MM] 6845 Determines the maximum allowed order for slabs. 6846 A high setting may cause OOMs due to memory 6847 fragmentation. For more information see 6848 Documentation/admin-guide/mm/slab.rst. 6849 (slub_max_order legacy name also accepted for now) 6850 6851 slab_merge [MM] 6852 Enable merging of slabs with similar size when the 6853 kernel is built without CONFIG_SLAB_MERGE_DEFAULT. 6854 (slub_merge legacy name also accepted for now) 6855 6856 slab_min_objects= [MM] 6857 The minimum number of objects per slab. SLUB will 6858 increase the slab order up to slab_max_order to 6859 generate a sufficiently large slab able to contain 6860 the number of objects indicated. The higher the number 6861 of objects the smaller the overhead of tracking slabs 6862 and the less frequently locks need to be acquired. 6863 For more information see 6864 Documentation/admin-guide/mm/slab.rst. 6865 (slub_min_objects legacy name also accepted for now) 6866 6867 slab_min_order= [MM] 6868 Determines the minimum page order for slabs. Must be 6869 lower or equal to slab_max_order. For more information see 6870 Documentation/admin-guide/mm/slab.rst. 6871 (slub_min_order legacy name also accepted for now) 6872 6873 slab_nomerge [MM] 6874 Disable merging of slabs with similar size. May be 6875 necessary if there is some reason to distinguish 6876 allocs to different slabs, especially in hardened 6877 environments where the risk of heap overflows and 6878 layout control by attackers can usually be 6879 frustrated by disabling merging. This will reduce 6880 most of the exposure of a heap attack to a single 6881 cache (risks via metadata attacks are mostly 6882 unchanged). Debug options disable merging on their 6883 own. 6884 For more information see 6885 Documentation/admin-guide/mm/slab.rst. 6886 (slub_nomerge legacy name also accepted for now) 6887 6888 slab_strict_numa [MM] 6889 Support memory policies on a per object level 6890 in the slab allocator. The default is for memory 6891 policies to be applied at the folio level when 6892 a new folio is needed or a partial folio is 6893 retrieved from the lists. Increases overhead 6894 in the slab fastpaths but gains more accurate 6895 NUMA kernel object placement which helps with slow 6896 interconnects in NUMA systems. 6897 6898 slram= [HW,MTD] 6899 6900 smart2= [HW] 6901 Format: <io1>[,<io2>[,...,<io8>]] 6902 6903 smp.csd_lock_timeout= [KNL] 6904 Specify the period of time in milliseconds 6905 that smp_call_function() and friends will wait 6906 for a CPU to release the CSD lock. This is 6907 useful when diagnosing bugs involving CPUs 6908 disabling interrupts for extended periods 6909 of time. Defaults to 5,000 milliseconds, and 6910 setting a value of zero disables this feature. 6911 This feature may be more efficiently disabled 6912 using the csdlock_debug- kernel parameter. 6913 6914 smp.panic_on_ipistall= [KNL] 6915 If a csd_lock_timeout extends for more than 6916 the specified number of milliseconds, panic the 6917 system. By default, let CSD-lock acquisition 6918 take as long as they take. Specifying 300,000 6919 for this value provides a 5-minute timeout. 6920 6921 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices 6922 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port 6923 smsc-ircc2.ircc_sir= [HW] SIR base I/O port 6924 smsc-ircc2.ircc_fir= [HW] FIR base I/O port 6925 smsc-ircc2.ircc_irq= [HW] IRQ line 6926 smsc-ircc2.ircc_dma= [HW] DMA channel 6927 smsc-ircc2.ircc_transceiver= [HW] Transceiver type: 6928 0: Toshiba Satellite 1800 (GP data pin select) 6929 1: Fast pin select (default) 6930 2: ATC IRMode 6931 6932 smt= [KNL,MIPS,S390,EARLY] Set the maximum number of threads 6933 (logical CPUs) to use per physical CPU on systems 6934 capable of symmetric multithreading (SMT). Will 6935 be capped to the actual hardware limit. 6936 Format: <integer> 6937 Default: -1 (no limit) 6938 6939 softlockup_panic= 6940 [KNL] Should the soft-lockup detector generate panics. 6941 Format: 0 | 1 6942 6943 A value of 1 instructs the soft-lockup detector 6944 to panic the machine when a soft-lockup occurs. It is 6945 also controlled by the kernel.softlockup_panic sysctl 6946 and CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC, which is the 6947 respective build-time switch to that functionality. 6948 6949 softlockup_all_cpu_backtrace= 6950 [KNL] Should the soft-lockup detector generate 6951 backtraces on all cpus. 6952 Format: 0 | 1 6953 6954 sonypi.*= [HW] Sony Programmable I/O Control Device driver 6955 See Documentation/admin-guide/laptops/sonypi.rst 6956 6957 spectre_bhi= [X86] Control mitigation of Branch History Injection 6958 (BHI) vulnerability. This setting affects the 6959 deployment of the HW BHI control and the SW BHB 6960 clearing sequence. 6961 6962 on - (default) Enable the HW or SW mitigation as 6963 needed. This protects the kernel from 6964 both syscalls and VMs. 6965 vmexit - On systems which don't have the HW mitigation 6966 available, enable the SW mitigation on vmexit 6967 ONLY. On such systems, the host kernel is 6968 protected from VM-originated BHI attacks, but 6969 may still be vulnerable to syscall attacks. 6970 off - Disable the mitigation. 6971 6972 spectre_v2= [X86,EARLY] Control mitigation of Spectre variant 2 6973 (indirect branch speculation) vulnerability. 6974 The default operation protects the kernel from 6975 user space attacks. 6976 6977 on - unconditionally enable, implies 6978 spectre_v2_user=on 6979 off - unconditionally disable, implies 6980 spectre_v2_user=off 6981 auto - kernel detects whether your CPU model is 6982 vulnerable 6983 6984 Selecting 'on' will, and 'auto' may, choose a 6985 mitigation method at run time according to the 6986 CPU, the available microcode, the setting of the 6987 CONFIG_MITIGATION_RETPOLINE configuration option, 6988 and the compiler with which the kernel was built. 6989 6990 Selecting 'on' will also enable the mitigation 6991 against user space to user space task attacks. 6992 Selecting specific mitigation does not force enable 6993 user mitigations. 6994 6995 Selecting 'off' will disable both the kernel and 6996 the user space protections. 6997 6998 Specific mitigations can also be selected manually: 6999 7000 retpoline - replace indirect branches 7001 retpoline,generic - Retpolines 7002 retpoline,lfence - LFENCE; indirect branch 7003 retpoline,amd - alias for retpoline,lfence 7004 eibrs - Enhanced/Auto IBRS 7005 eibrs,retpoline - Enhanced/Auto IBRS + Retpolines 7006 eibrs,lfence - Enhanced/Auto IBRS + LFENCE 7007 ibrs - use IBRS to protect kernel 7008 7009 Not specifying this option is equivalent to 7010 spectre_v2=auto. 7011 7012 spectre_v2_user= 7013 [X86] Control mitigation of Spectre variant 2 7014 (indirect branch speculation) vulnerability between 7015 user space tasks 7016 7017 on - Unconditionally enable mitigations. Is 7018 enforced by spectre_v2=on 7019 7020 off - Unconditionally disable mitigations. Is 7021 enforced by spectre_v2=off 7022 7023 prctl - Indirect branch speculation is enabled, 7024 but mitigation can be enabled via prctl 7025 per thread. The mitigation control state 7026 is inherited on fork. 7027 7028 prctl,ibpb 7029 - Like "prctl" above, but only STIBP is 7030 controlled per thread. IBPB is issued 7031 always when switching between different user 7032 space processes. 7033 7034 seccomp 7035 - Same as "prctl" above, but all seccomp 7036 threads will enable the mitigation unless 7037 they explicitly opt out. 7038 7039 seccomp,ibpb 7040 - Like "seccomp" above, but only STIBP is 7041 controlled per thread. IBPB is issued 7042 always when switching between different 7043 user space processes. 7044 7045 auto - Kernel selects the mitigation depending on 7046 the available CPU features and vulnerability. 7047 7048 Default mitigation: "prctl" 7049 7050 Not specifying this option is equivalent to 7051 spectre_v2_user=auto. 7052 7053 spec_rstack_overflow= 7054 [X86,EARLY] Control RAS overflow mitigation on AMD Zen CPUs 7055 7056 off - Disable mitigation 7057 microcode - Enable microcode mitigation only 7058 safe-ret - Enable sw-only safe RET mitigation (default) 7059 ibpb - Enable mitigation by issuing IBPB on 7060 kernel entry 7061 ibpb-vmexit - Issue IBPB only on VMEXIT 7062 (cloud-specific mitigation) 7063 7064 spec_store_bypass_disable= 7065 [HW,EARLY] Control Speculative Store Bypass (SSB) Disable mitigation 7066 (Speculative Store Bypass vulnerability) 7067 7068 Certain CPUs are vulnerable to an exploit against a 7069 a common industry wide performance optimization known 7070 as "Speculative Store Bypass" in which recent stores 7071 to the same memory location may not be observed by 7072 later loads during speculative execution. The idea 7073 is that such stores are unlikely and that they can 7074 be detected prior to instruction retirement at the 7075 end of a particular speculation execution window. 7076 7077 In vulnerable processors, the speculatively forwarded 7078 store can be used in a cache side channel attack, for 7079 example to read memory to which the attacker does not 7080 directly have access (e.g. inside sandboxed code). 7081 7082 This parameter controls whether the Speculative Store 7083 Bypass optimization is used. 7084 7085 On x86 the options are: 7086 7087 on - Unconditionally disable Speculative Store Bypass 7088 off - Unconditionally enable Speculative Store Bypass 7089 auto - Kernel detects whether the CPU model contains an 7090 implementation of Speculative Store Bypass and 7091 picks the most appropriate mitigation. If the 7092 CPU is not vulnerable, "off" is selected. If the 7093 CPU is vulnerable the default mitigation is 7094 architecture and Kconfig dependent. See below. 7095 prctl - Control Speculative Store Bypass per thread 7096 via prctl. Speculative Store Bypass is enabled 7097 for a process by default. The state of the control 7098 is inherited on fork. 7099 seccomp - Same as "prctl" above, but all seccomp threads 7100 will disable SSB unless they explicitly opt out. 7101 7102 Default mitigations: 7103 X86: "prctl" 7104 7105 On powerpc the options are: 7106 7107 on,auto - On Power8 and Power9 insert a store-forwarding 7108 barrier on kernel entry and exit. On Power7 7109 perform a software flush on kernel entry and 7110 exit. 7111 off - No action. 7112 7113 Not specifying this option is equivalent to 7114 spec_store_bypass_disable=auto. 7115 7116 split_lock_detect= 7117 [X86] Enable split lock detection or bus lock detection 7118 7119 When enabled (and if hardware support is present), atomic 7120 instructions that access data across cache line 7121 boundaries will result in an alignment check exception 7122 for split lock detection or a debug exception for 7123 bus lock detection. 7124 7125 off - not enabled 7126 7127 warn - the kernel will emit rate-limited warnings 7128 about applications triggering the #AC 7129 exception or the #DB exception. This mode is 7130 the default on CPUs that support split lock 7131 detection or bus lock detection. Default 7132 behavior is by #AC if both features are 7133 enabled in hardware. 7134 7135 fatal - the kernel will send SIGBUS to applications 7136 that trigger the #AC exception or the #DB 7137 exception. Default behavior is by #AC if 7138 both features are enabled in hardware. 7139 7140 ratelimit:N - 7141 Set system wide rate limit to N bus locks 7142 per second for bus lock detection. 7143 0 < N <= 1000. 7144 7145 N/A for split lock detection. 7146 7147 7148 If an #AC exception is hit in the kernel or in 7149 firmware (i.e. not while executing in user mode) 7150 the kernel will oops in either "warn" or "fatal" 7151 mode. 7152 7153 #DB exception for bus lock is triggered only when 7154 CPL > 0. 7155 7156 srbds= [X86,INTEL,EARLY] 7157 Control the Special Register Buffer Data Sampling 7158 (SRBDS) mitigation. 7159 7160 Certain CPUs are vulnerable to an MDS-like 7161 exploit which can leak bits from the random 7162 number generator. 7163 7164 By default, this issue is mitigated by 7165 microcode. However, the microcode fix can cause 7166 the RDRAND and RDSEED instructions to become 7167 much slower. Among other effects, this will 7168 result in reduced throughput from /dev/urandom. 7169 7170 The microcode mitigation can be disabled with 7171 the following option: 7172 7173 off: Disable mitigation and remove 7174 performance impact to RDRAND and RDSEED 7175 7176 srcutree.big_cpu_lim [KNL] 7177 Specifies the number of CPUs constituting a 7178 large system, such that srcu_struct structures 7179 should immediately allocate an srcu_node array. 7180 This kernel-boot parameter defaults to 128, 7181 but takes effect only when the low-order four 7182 bits of srcutree.convert_to_big is equal to 3 7183 (decide at boot). 7184 7185 srcutree.convert_to_big [KNL] 7186 Specifies under what conditions an SRCU tree 7187 srcu_struct structure will be converted to big 7188 form, that is, with an rcu_node tree: 7189 7190 0: Never. 7191 1: At init_srcu_struct() time. 7192 2: When rcutorture decides to. 7193 3: Decide at boot time (default). 7194 0x1X: Above plus if high contention. 7195 7196 Either way, the srcu_node tree will be sized based 7197 on the actual runtime number of CPUs (nr_cpu_ids) 7198 instead of the compile-time CONFIG_NR_CPUS. 7199 7200 srcutree.counter_wrap_check [KNL] 7201 Specifies how frequently to check for 7202 grace-period sequence counter wrap for the 7203 srcu_data structure's ->srcu_gp_seq_needed field. 7204 The greater the number of bits set in this kernel 7205 parameter, the less frequently counter wrap will 7206 be checked for. Note that the bottom two bits 7207 are ignored. 7208 7209 srcutree.exp_holdoff [KNL] 7210 Specifies how many nanoseconds must elapse 7211 since the end of the last SRCU grace period for 7212 a given srcu_struct until the next normal SRCU 7213 grace period will be considered for automatic 7214 expediting. Set to zero to disable automatic 7215 expediting. 7216 7217 srcutree.srcu_max_nodelay [KNL] 7218 Specifies the number of no-delay instances 7219 per jiffy for which the SRCU grace period 7220 worker thread will be rescheduled with zero 7221 delay. Beyond this limit, worker thread will 7222 be rescheduled with a sleep delay of one jiffy. 7223 7224 srcutree.srcu_max_nodelay_phase [KNL] 7225 Specifies the per-grace-period phase, number of 7226 non-sleeping polls of readers. Beyond this limit, 7227 grace period worker thread will be rescheduled 7228 with a sleep delay of one jiffy, between each 7229 rescan of the readers, for a grace period phase. 7230 7231 srcutree.srcu_retry_check_delay [KNL] 7232 Specifies number of microseconds of non-sleeping 7233 delay between each non-sleeping poll of readers. 7234 7235 srcutree.small_contention_lim [KNL] 7236 Specifies the number of update-side contention 7237 events per jiffy will be tolerated before 7238 initiating a conversion of an srcu_struct 7239 structure to big form. Note that the value of 7240 srcutree.convert_to_big must have the 0x10 bit 7241 set for contention-based conversions to occur. 7242 7243 ssbd= [ARM64,HW,EARLY] 7244 Speculative Store Bypass Disable control 7245 7246 On CPUs that are vulnerable to the Speculative 7247 Store Bypass vulnerability and offer a 7248 firmware based mitigation, this parameter 7249 indicates how the mitigation should be used: 7250 7251 force-on: Unconditionally enable mitigation for 7252 for both kernel and userspace 7253 force-off: Unconditionally disable mitigation for 7254 for both kernel and userspace 7255 kernel: Always enable mitigation in the 7256 kernel, and offer a prctl interface 7257 to allow userspace to register its 7258 interest in being mitigated too. 7259 7260 stack_guard_gap= [MM] 7261 override the default stack gap protection. The value 7262 is in page units and it defines how many pages prior 7263 to (for stacks growing down) resp. after (for stacks 7264 growing up) the main stack are reserved for no other 7265 mapping. Default value is 256 pages. 7266 7267 stack_depot_disable= [KNL,EARLY] 7268 Setting this to true through kernel command line will 7269 disable the stack depot thereby saving the static memory 7270 consumed by the stack hash table. By default this is set 7271 to false. 7272 7273 stack_depot_max_pools= [KNL,EARLY] 7274 Specify the maximum number of pools to use for storing 7275 stack traces. Pools are allocated on-demand up to this 7276 limit. Default value is 8191 pools. 7277 7278 stacktrace [FTRACE] 7279 Enable the stack tracer on boot up. 7280 7281 stacktrace_filter=[function-list] 7282 [FTRACE] Limit the functions that the stack tracer 7283 will trace at boot up. function-list is a comma-separated 7284 list of functions. This list can be changed at run 7285 time by the stack_trace_filter file in the debugfs 7286 tracing directory. Note, this enables stack tracing 7287 and the stacktrace above is not needed. 7288 7289 sti= [PARISC,HW] 7290 Format: <num> 7291 Set the STI (builtin display/keyboard on the HP-PARISC 7292 machines) console (graphic card) which should be used 7293 as the initial boot-console. 7294 See also comment in drivers/video/console/sticore.c. 7295 7296 sti_font= [HW] 7297 See comment in drivers/video/console/sticore.c. 7298 7299 stifb= [HW] 7300 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]] 7301 7302 strict_sas_size= 7303 [X86] 7304 Format: <bool> 7305 Enable or disable strict sigaltstack size checks 7306 against the required signal frame size which 7307 depends on the supported FPU features. This can 7308 be used to filter out binaries which have 7309 not yet been made aware of AT_MINSIGSTKSZ. 7310 7311 stress_hpt [PPC,EARLY] 7312 Limits the number of kernel HPT entries in the hash 7313 page table to increase the rate of hash page table 7314 faults on kernel addresses. 7315 7316 stress_slb [PPC,EARLY] 7317 Limits the number of kernel SLB entries, and flushes 7318 them frequently to increase the rate of SLB faults 7319 on kernel addresses. 7320 7321 no_slb_preload [PPC,EARLY] 7322 Disables slb preloading for userspace. 7323 7324 sunrpc.min_resvport= 7325 sunrpc.max_resvport= 7326 [NFS,SUNRPC] 7327 SunRPC servers often require that client requests 7328 originate from a privileged port (i.e. a port in the 7329 range 0 < portnr < 1024). 7330 An administrator who wishes to reserve some of these 7331 ports for other uses may adjust the range that the 7332 kernel's sunrpc client considers to be privileged 7333 using these two parameters to set the minimum and 7334 maximum port values. 7335 7336 sunrpc.svc_rpc_per_connection_limit= 7337 [NFS,SUNRPC] 7338 Limit the number of requests that the server will 7339 process in parallel from a single connection. 7340 The default value is 0 (no limit). 7341 7342 sunrpc.pool_mode= 7343 [NFS] 7344 Control how the NFS server code allocates CPUs to 7345 service thread pools. Depending on how many NICs 7346 you have and where their interrupts are bound, this 7347 option will affect which CPUs will do NFS serving. 7348 Note: this parameter cannot be changed while the 7349 NFS server is running. 7350 7351 auto the server chooses an appropriate mode 7352 automatically using heuristics 7353 global a single global pool contains all CPUs 7354 percpu one pool for each CPU 7355 pernode one pool for each NUMA node (equivalent 7356 to global on non-NUMA machines) 7357 7358 sunrpc.tcp_slot_table_entries= 7359 sunrpc.udp_slot_table_entries= 7360 [NFS,SUNRPC] 7361 Sets the upper limit on the number of simultaneous 7362 RPC calls that can be sent from the client to a 7363 server. Increasing these values may allow you to 7364 improve throughput, but will also increase the 7365 amount of memory reserved for use by the client. 7366 7367 suspend.pm_test_delay= 7368 [SUSPEND] 7369 Sets the number of seconds to remain in a suspend test 7370 mode before resuming the system (see 7371 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG 7372 is set. Default value is 5. 7373 7374 svm= [PPC] 7375 Format: { on | off | y | n | 1 | 0 } 7376 This parameter controls use of the Protected 7377 Execution Facility on pSeries. 7378 7379 swiotlb= [ARM,PPC,MIPS,X86,S390,EARLY] 7380 Format: { <int> [,<int>] | force | noforce } 7381 <int> -- Number of I/O TLB slabs 7382 <int> -- Second integer after comma. Number of swiotlb 7383 areas with their own lock. Will be rounded up 7384 to a power of 2. 7385 force -- force using of bounce buffers even if they 7386 wouldn't be automatically used by the kernel 7387 noforce -- Never use bounce buffers (for debugging) 7388 7389 switches= [HW,M68k,EARLY] 7390 7391 sysctl.*= [KNL] 7392 Set a sysctl parameter, right before loading the init 7393 process, as if the value was written to the respective 7394 /proc/sys/... file. Both '.' and '/' are recognized as 7395 separators. Unrecognized parameters and invalid values 7396 are reported in the kernel log. Sysctls registered 7397 later by a loaded module cannot be set this way. 7398 Example: sysctl.vm.swappiness=40 7399 7400 sysrq_always_enabled 7401 [KNL] 7402 Ignore sysrq setting - this boot parameter will 7403 neutralize any effect of /proc/sys/kernel/sysrq. 7404 Useful for debugging. 7405 7406 tcpmhash_entries= [KNL,NET] 7407 Set the number of tcp_metrics_hash slots. 7408 Default value is 8192 or 16384 depending on total 7409 ram pages. This is used to specify the TCP metrics 7410 cache size. See Documentation/networking/ip-sysctl.rst 7411 "tcp_no_metrics_save" section for more details. 7412 7413 tdfx= [HW,DRM] 7414 7415 test_suspend= [SUSPEND] 7416 Format: { "mem" | "standby" | "freeze" }[,N] 7417 Specify "mem" (for Suspend-to-RAM) or "standby" (for 7418 standby suspend) or "freeze" (for suspend type freeze) 7419 as the system sleep state during system startup with 7420 the optional capability to repeat N number of times. 7421 The system is woken from this state using a 7422 wakeup-capable RTC alarm. 7423 7424 thash_entries= [KNL,NET] 7425 Set number of hash buckets for TCP connection 7426 7427 thermal.act= [HW,ACPI] 7428 -1: disable all active trip points in all thermal zones 7429 <degrees C>: override all lowest active trip points 7430 7431 thermal.crt= [HW,ACPI] 7432 -1: disable all critical trip points in all thermal zones 7433 <degrees C>: override all critical trip points 7434 7435 thermal.off= [HW,ACPI] 7436 1: disable ACPI thermal control 7437 7438 thermal.psv= [HW,ACPI] 7439 -1: disable all passive trip points 7440 <degrees C>: override all passive trip points to this 7441 value 7442 7443 thermal.tzp= [HW,ACPI] 7444 Specify global default ACPI thermal zone polling rate 7445 <deci-seconds>: poll all this frequency 7446 0: no polling (default) 7447 7448 thp_anon= [KNL] 7449 Format: <size>[KMG],<size>[KMG]:<state>;<size>[KMG]-<size>[KMG]:<state> 7450 state is one of "always", "madvise", "never" or "inherit". 7451 Control the default behavior of the system with respect 7452 to anonymous transparent hugepages. 7453 Can be used multiple times for multiple anon THP sizes. 7454 See Documentation/admin-guide/mm/transhuge.rst for more 7455 details. 7456 7457 threadirqs [KNL,EARLY] 7458 Force threading of all interrupt handlers except those 7459 marked explicitly IRQF_NO_THREAD. 7460 7461 thp_shmem= [KNL] 7462 Format: <size>[KMG],<size>[KMG]:<policy>;<size>[KMG]-<size>[KMG]:<policy> 7463 Control the default policy of each hugepage size for the 7464 internal shmem mount. <policy> is one of policies available 7465 for the shmem mount ("always", "inherit", "never", "within_size", 7466 and "advise"). 7467 It can be used multiple times for multiple shmem THP sizes. 7468 See Documentation/admin-guide/mm/transhuge.rst for more 7469 details. 7470 7471 topology= [S390,EARLY] 7472 Format: {off | on} 7473 Specify if the kernel should make use of the cpu 7474 topology information if the hardware supports this. 7475 The scheduler will make use of this information and 7476 e.g. base its process migration decisions on it. 7477 Default is on. 7478 7479 torture.disable_onoff_at_boot= [KNL] 7480 Prevent the CPU-hotplug component of torturing 7481 until after init has spawned. 7482 7483 torture.ftrace_dump_at_shutdown= [KNL] 7484 Dump the ftrace buffer at torture-test shutdown, 7485 even if there were no errors. This can be a 7486 very costly operation when many torture tests 7487 are running concurrently, especially on systems 7488 with rotating-rust storage. 7489 7490 torture.verbose_sleep_frequency= [KNL] 7491 Specifies how many verbose printk()s should be 7492 emitted between each sleep. The default of zero 7493 disables verbose-printk() sleeping. 7494 7495 torture.verbose_sleep_duration= [KNL] 7496 Duration of each verbose-printk() sleep in jiffies. 7497 7498 tpm.disable_pcr_integrity= [HW,TPM] 7499 Do not protect PCR registers from unintended physical 7500 access, or interposers in the bus by the means of 7501 having an integrity protected session wrapped around 7502 TPM2_PCR_Extend command. Consider this in a situation 7503 where TPM is heavily utilized by IMA, thus protection 7504 causing a major performance hit, and the space where 7505 machines are deployed is by other means guarded. 7506 7507 tpm_crb_ffa.busy_timeout_ms= [ARM64,TPM] 7508 Maximum time in milliseconds to retry sending a message 7509 to the TPM service before giving up. This parameter controls 7510 how long the system will continue retrying when the TPM 7511 service is busy. 7512 Format: <unsigned int> 7513 Default: 2000 (2 seconds) 7514 7515 tpm_suspend_pcr=[HW,TPM] 7516 Format: integer pcr id 7517 Specify that at suspend time, the tpm driver 7518 should extend the specified pcr with zeros, 7519 as a workaround for some chips which fail to 7520 flush the last written pcr on TPM_SaveState. 7521 This will guarantee that all the other pcrs 7522 are saved. 7523 7524 tpm_tis.interrupts= [HW,TPM] 7525 Enable interrupts for the MMIO based physical layer 7526 for the FIFO interface. By default it is set to false 7527 (0). For more information about TPM hardware interfaces 7528 defined by Trusted Computing Group (TCG) see 7529 https://trustedcomputinggroup.org/resource/pc-client-platform-tpm-profile-ptp-specification/ 7530 7531 tp_printk [FTRACE] 7532 Have the tracepoints sent to printk as well as the 7533 tracing ring buffer. This is useful for early boot up 7534 where the system hangs or reboots and does not give the 7535 option for reading the tracing buffer or performing a 7536 ftrace_dump_on_oops. 7537 7538 To turn off having tracepoints sent to printk, 7539 echo 0 > /proc/sys/kernel/tracepoint_printk 7540 Note, echoing 1 into this file without the 7541 tp_printk kernel cmdline option has no effect. 7542 7543 The tp_printk_stop_on_boot (see below) can also be used 7544 to stop the printing of events to console at 7545 late_initcall_sync. 7546 7547 ** CAUTION ** 7548 7549 Having tracepoints sent to printk() and activating high 7550 frequency tracepoints such as irq or sched, can cause 7551 the system to live lock. 7552 7553 tp_printk_stop_on_boot [FTRACE] 7554 When tp_printk (above) is set, it can cause a lot of noise 7555 on the console. It may be useful to only include the 7556 printing of events during boot up, as user space may 7557 make the system inoperable. 7558 7559 This command line option will stop the printing of events 7560 to console at the late_initcall_sync() time frame. 7561 7562 trace_buf_size=nn[KMG] 7563 [FTRACE] will set tracing buffer size on each cpu. 7564 7565 trace_clock= [FTRACE] Set the clock used for tracing events 7566 at boot up. 7567 local - Use the per CPU time stamp counter 7568 (converted into nanoseconds). Fast, but 7569 depending on the architecture, may not be 7570 in sync between CPUs. 7571 global - Event time stamps are synchronized across 7572 CPUs. May be slower than the local clock, 7573 but better for some race conditions. 7574 counter - Simple counting of events (1, 2, ..) 7575 note, some counts may be skipped due to the 7576 infrastructure grabbing the clock more than 7577 once per event. 7578 uptime - Use jiffies as the time stamp. 7579 perf - Use the same clock that perf uses. 7580 mono - Use ktime_get_mono_fast_ns() for time stamps. 7581 mono_raw - Use ktime_get_raw_fast_ns() for time 7582 stamps. 7583 boot - Use ktime_get_boot_fast_ns() for time stamps. 7584 Architectures may add more clocks. See 7585 Documentation/trace/ftrace.rst for more details. 7586 7587 trace_event=[event-list] 7588 [FTRACE] Set and start specified trace events in order 7589 to facilitate early boot debugging. The event-list is a 7590 comma-separated list of trace events to enable. See 7591 also Documentation/trace/events.rst 7592 7593 To enable modules, use :mod: keyword: 7594 7595 trace_event=:mod:<module> 7596 7597 The value before :mod: will only enable specific events 7598 that are part of the module. See the above mentioned 7599 document for more information. 7600 7601 trace_instance=[instance-info] 7602 [FTRACE] Create a ring buffer instance early in boot up. 7603 This will be listed in: 7604 7605 /sys/kernel/tracing/instances 7606 7607 Events can be enabled at the time the instance is created 7608 via: 7609 7610 trace_instance=<name>,<system1>:<event1>,<system2>:<event2> 7611 7612 Note, the "<system*>:" portion is optional if the event is 7613 unique. 7614 7615 trace_instance=foo,sched:sched_switch,irq_handler_entry,initcall 7616 7617 will enable the "sched_switch" event (note, the "sched:" is optional, and 7618 the same thing would happen if it was left off). The irq_handler_entry 7619 event, and all events under the "initcall" system. 7620 7621 Flags can be added to the instance to modify its behavior when it is 7622 created. The flags are separated by '^'. 7623 7624 The available flags are: 7625 7626 traceoff - Have the tracing instance tracing disabled after it is created. 7627 traceprintk - Have trace_printk() write into this trace instance 7628 (note, "printk" and "trace_printk" can also be used) 7629 7630 trace_instance=foo^traceoff^traceprintk,sched,irq 7631 7632 The flags must come before the defined events. 7633 7634 If memory has been reserved (see memmap for x86), the instance 7635 can use that memory: 7636 7637 memmap=12M$0x284500000 trace_instance=boot_map@0x284500000:12M 7638 7639 The above will create a "boot_map" instance that uses the physical 7640 memory at 0x284500000 that is 12Megs. The per CPU buffers of that 7641 instance will be split up accordingly. 7642 7643 Alternatively, the memory can be reserved by the reserve_mem option: 7644 7645 reserve_mem=12M:4096:trace trace_instance=boot_map@trace 7646 7647 This will reserve 12 megabytes at boot up with a 4096 byte alignment 7648 and place the ring buffer in this memory. Note that due to KASLR, the 7649 memory may not be the same location each time, which will not preserve 7650 the buffer content. 7651 7652 Also note that the layout of the ring buffer data may change between 7653 kernel versions where the validator will fail and reset the ring buffer 7654 if the layout is not the same as the previous kernel. 7655 7656 If the ring buffer is used for persistent bootups and has events enabled, 7657 it is recommend to disable tracing so that events from a previous boot do not 7658 mix with events of the current boot (unless you are debugging a random crash 7659 at boot up). 7660 7661 reserve_mem=12M:4096:trace trace_instance=boot_map^traceoff^traceprintk@trace,sched,irq 7662 7663 Note, saving the trace buffer across reboots does require that the system 7664 is set up to not wipe memory. For instance, CONFIG_RESET_ATTACK_MITIGATION 7665 can force a memory reset on boot which will clear any trace that was stored. 7666 This is just one of many ways that can clear memory. Make sure your system 7667 keeps the content of memory across reboots before relying on this option. 7668 7669 NB: Both the mapped address and size must be page aligned for the architecture. 7670 7671 See also Documentation/trace/debugging.rst 7672 7673 7674 trace_options=[option-list] 7675 [FTRACE] Enable or disable tracer options at boot. 7676 The option-list is a comma delimited list of options 7677 that can be enabled or disabled just as if you were 7678 to echo the option name into 7679 7680 /sys/kernel/tracing/trace_options 7681 7682 For example, to enable stacktrace option (to dump the 7683 stack trace of each event), add to the command line: 7684 7685 trace_options=stacktrace 7686 7687 See also Documentation/trace/ftrace.rst "trace options" 7688 section. 7689 7690 trace_trigger=[trigger-list] 7691 [FTRACE] Add an event trigger on specific events. 7692 Set a trigger on top of a specific event, with an optional 7693 filter. 7694 7695 The format is "trace_trigger=<event>.<trigger>[ if <filter>],..." 7696 Where more than one trigger may be specified that are comma delimited. 7697 7698 For example: 7699 7700 trace_trigger="sched_switch.stacktrace if prev_state == 2" 7701 7702 The above will enable the "stacktrace" trigger on the "sched_switch" 7703 event but only trigger it if the "prev_state" of the "sched_switch" 7704 event is "2" (TASK_UNINTERRUPTIBLE). 7705 7706 See also "Event triggers" in Documentation/trace/events.rst 7707 7708 7709 traceoff_after_boot 7710 [FTRACE] Sometimes tracing is used to debug issues 7711 during the boot process. Since the trace buffer has a 7712 limited amount of storage, it may be prudent to 7713 disable tracing after the boot is finished, otherwise 7714 the critical information may be overwritten. With this 7715 option, the main tracing buffer will be turned off at 7716 the end of the boot process. 7717 7718 traceoff_on_warning 7719 [FTRACE] enable this option to disable tracing when a 7720 warning is hit. This turns off "tracing_on". Tracing can 7721 be enabled again by echoing '1' into the "tracing_on" 7722 file located in /sys/kernel/tracing/ 7723 7724 This option is useful, as it disables the trace before 7725 the WARNING dump is called, which prevents the trace to 7726 be filled with content caused by the warning output. 7727 7728 This option can also be set at run time via the sysctl 7729 option: kernel/traceoff_on_warning 7730 7731 transparent_hugepage= 7732 [KNL] 7733 Format: [always|madvise|never] 7734 Can be used to control the default behavior of the system 7735 with respect to transparent hugepages. 7736 See Documentation/admin-guide/mm/transhuge.rst 7737 for more details. 7738 7739 transparent_hugepage_shmem= [KNL] 7740 Format: [always|within_size|advise|never|deny|force] 7741 Can be used to control the hugepage allocation policy for 7742 the internal shmem mount. 7743 See Documentation/admin-guide/mm/transhuge.rst 7744 for more details. 7745 7746 transparent_hugepage_tmpfs= [KNL] 7747 Format: [always|within_size|advise|never] 7748 Can be used to control the default hugepage allocation policy 7749 for the tmpfs mount. 7750 See Documentation/admin-guide/mm/transhuge.rst 7751 for more details. 7752 7753 trusted.source= [KEYS] 7754 Format: <string> 7755 This parameter identifies the trust source as a backend 7756 for trusted keys implementation. Supported trust 7757 sources: 7758 - "tpm" 7759 - "tee" 7760 - "caam" 7761 - "dcp" 7762 If not specified then it defaults to iterating through 7763 the trust source list starting with TPM and assigns the 7764 first trust source as a backend which is initialized 7765 successfully during iteration. 7766 7767 trusted.rng= [KEYS] 7768 Format: <string> 7769 The RNG used to generate key material for trusted keys. 7770 Can be one of: 7771 - "kernel" 7772 - the same value as trusted.source: "tpm" or "tee" 7773 - "default" 7774 If not specified, "default" is used. In this case, 7775 the RNG's choice is left to each individual trust source. 7776 7777 trusted.dcp_use_otp_key 7778 This is intended to be used in combination with 7779 trusted.source=dcp and will select the DCP OTP key 7780 instead of the DCP UNIQUE key blob encryption. 7781 7782 trusted.dcp_skip_zk_test 7783 This is intended to be used in combination with 7784 trusted.source=dcp and will disable the check if the 7785 blob key is all zeros. This is helpful for situations where 7786 having this key zero'ed is acceptable. E.g. in testing 7787 scenarios. 7788 7789 tsa= [X86] Control mitigation for Transient Scheduler 7790 Attacks on AMD CPUs. Search the following in your 7791 favourite search engine for more details: 7792 7793 "Technical guidance for mitigating transient scheduler 7794 attacks". 7795 7796 off - disable the mitigation 7797 on - enable the mitigation (default) 7798 user - mitigate only user/kernel transitions 7799 vm - mitigate only guest/host transitions 7800 7801 7802 tsc= Disable clocksource stability checks for TSC. 7803 Format: <string> 7804 [x86] reliable: mark tsc clocksource as reliable, this 7805 disables clocksource verification at runtime, as well 7806 as the stability checks done at bootup. Used to enable 7807 high-resolution timer mode on older hardware, and in 7808 virtualized environment. 7809 [x86] noirqtime: Do not use TSC to do irq accounting. 7810 Used to run time disable IRQ_TIME_ACCOUNTING on any 7811 platforms where RDTSC is slow and this accounting 7812 can add overhead. 7813 [x86] unstable: mark the TSC clocksource as unstable, this 7814 marks the TSC unconditionally unstable at bootup and 7815 avoids any further wobbles once the TSC watchdog notices. 7816 [x86] nowatchdog: disable clocksource watchdog. Used 7817 in situations with strict latency requirements (where 7818 interruptions from clocksource watchdog are not 7819 acceptable). 7820 [x86] recalibrate: force recalibration against a HW timer 7821 (HPET or PM timer) on systems whose TSC frequency was 7822 obtained from HW or FW using either an MSR or CPUID(0x15). 7823 Warn if the difference is more than 500 ppm. 7824 [x86] watchdog: Use TSC as the watchdog clocksource with 7825 which to check other HW timers (HPET or PM timer), but 7826 only on systems where TSC has been deemed trustworthy. 7827 This will be suppressed by an earlier tsc=nowatchdog and 7828 can be overridden by a later tsc=nowatchdog. A console 7829 message will flag any such suppression or overriding. 7830 7831 tsc_early_khz= [X86,EARLY] Skip early TSC calibration and use the given 7832 value instead. Useful when the early TSC frequency discovery 7833 procedure is not reliable, such as on overclocked systems 7834 with CPUID.16h support and partial CPUID.15h support. 7835 Format: <unsigned int> 7836 7837 tsx= [X86] Control Transactional Synchronization 7838 Extensions (TSX) feature in Intel processors that 7839 support TSX control. 7840 7841 This parameter controls the TSX feature. The options are: 7842 7843 on - Enable TSX on the system. Although there are 7844 mitigations for all known security vulnerabilities, 7845 TSX has been known to be an accelerator for 7846 several previous speculation-related CVEs, and 7847 so there may be unknown security risks associated 7848 with leaving it enabled. 7849 7850 off - Disable TSX on the system. (Note that this 7851 option takes effect only on newer CPUs which are 7852 not vulnerable to MDS, i.e., have 7853 MSR_IA32_ARCH_CAPABILITIES.MDS_NO=1 and which get 7854 the new IA32_TSX_CTRL MSR through a microcode 7855 update. This new MSR allows for the reliable 7856 deactivation of the TSX functionality.) 7857 7858 auto - Disable TSX if X86_BUG_TAA is present, 7859 otherwise enable TSX on the system. 7860 7861 Not specifying this option is equivalent to tsx=off. 7862 7863 See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst 7864 for more details. 7865 7866 tsx_async_abort= [X86,INTEL,EARLY] Control mitigation for the TSX Async 7867 Abort (TAA) vulnerability. 7868 7869 Similar to Micro-architectural Data Sampling (MDS) 7870 certain CPUs that support Transactional 7871 Synchronization Extensions (TSX) are vulnerable to an 7872 exploit against CPU internal buffers which can forward 7873 information to a disclosure gadget under certain 7874 conditions. 7875 7876 In vulnerable processors, the speculatively forwarded 7877 data can be used in a cache side channel attack, to 7878 access data to which the attacker does not have direct 7879 access. 7880 7881 This parameter controls the TAA mitigation. The 7882 options are: 7883 7884 full - Enable TAA mitigation on vulnerable CPUs 7885 if TSX is enabled. 7886 7887 full,nosmt - Enable TAA mitigation and disable SMT on 7888 vulnerable CPUs. If TSX is disabled, SMT 7889 is not disabled because CPU is not 7890 vulnerable to cross-thread TAA attacks. 7891 off - Unconditionally disable TAA mitigation 7892 7893 On MDS-affected machines, tsx_async_abort=off can be 7894 prevented by an active MDS mitigation as both vulnerabilities 7895 are mitigated with the same mechanism so in order to disable 7896 this mitigation, you need to specify mds=off too. 7897 7898 Not specifying this option is equivalent to 7899 tsx_async_abort=full. On CPUs which are MDS affected 7900 and deploy MDS mitigation, TAA mitigation is not 7901 required and doesn't provide any additional 7902 mitigation. 7903 7904 For details see: 7905 Documentation/admin-guide/hw-vuln/tsx_async_abort.rst 7906 7907 turbografx.map[2|3]= [HW,JOY] 7908 TurboGraFX parallel port interface 7909 Format: 7910 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7> 7911 See also Documentation/input/devices/joystick-parport.rst 7912 7913 udbg-immortal [PPC] When debugging early kernel crashes that 7914 happen after console_init() and before a proper 7915 console driver takes over, this boot options might 7916 help "seeing" what's going on. 7917 7918 uhash_entries= [KNL,NET] 7919 Set number of hash buckets for UDP/UDP-Lite connections 7920 7921 uhci-hcd.ignore_oc= 7922 [USB] Ignore overcurrent events (default N). 7923 Some badly-designed motherboards generate lots of 7924 bogus events, for ports that aren't wired to 7925 anything. Set this parameter to avoid log spamming. 7926 Note that genuine overcurrent events won't be 7927 reported either. 7928 7929 unaligned_scalar_speed= 7930 [RISCV] 7931 Format: {slow | fast | unsupported} 7932 Allow skipping scalar unaligned access speed tests. This 7933 is useful for testing alternative code paths and to skip 7934 the tests in environments where they run too slowly. All 7935 CPUs must have the same scalar unaligned access speed. 7936 7937 unaligned_vector_speed= 7938 [RISCV] 7939 Format: {slow | fast | unsupported} 7940 Allow skipping vector unaligned access speed tests. This 7941 is useful for testing alternative code paths and to skip 7942 the tests in environments where they run too slowly. All 7943 CPUs must have the same vector unaligned access speed. 7944 7945 unknown_nmi_panic 7946 [X86] Cause panic on unknown NMI. 7947 7948 unwind_debug [X86-64,EARLY] 7949 Enable unwinder debug output. This can be 7950 useful for debugging certain unwinder error 7951 conditions, including corrupt stacks and 7952 bad/missing unwinder metadata. 7953 7954 usbcore.authorized_default= 7955 [USB] Default USB device authorization: 7956 (default -1 = authorized (same as 1), 7957 0 = not authorized, 1 = authorized, 2 = authorized 7958 if device connected to internal port) 7959 7960 usbcore.autosuspend= 7961 [USB] The autosuspend time delay (in seconds) used 7962 for newly-detected USB devices (default 2). This 7963 is the time required before an idle device will be 7964 autosuspended. Devices for which the delay is set 7965 to a negative value won't be autosuspended at all. 7966 7967 usbcore.usbfs_snoop= 7968 [USB] Set to log all usbfs traffic (default 0 = off). 7969 7970 usbcore.usbfs_snoop_max= 7971 [USB] Maximum number of bytes to snoop in each URB 7972 (default = 65536). 7973 7974 usbcore.blinkenlights= 7975 [USB] Set to cycle leds on hubs (default 0 = off). 7976 7977 usbcore.old_scheme_first= 7978 [USB] Start with the old device initialization 7979 scheme (default 0 = off). 7980 7981 usbcore.usbfs_memory_mb= 7982 [USB] Memory limit (in MB) for buffers allocated by 7983 usbfs (default = 16, 0 = max = 2047). 7984 7985 usbcore.use_both_schemes= 7986 [USB] Try the other device initialization scheme 7987 if the first one fails (default 1 = enabled). 7988 7989 usbcore.initial_descriptor_timeout= 7990 [USB] Specifies timeout for the initial 64-byte 7991 USB_REQ_GET_DESCRIPTOR request in milliseconds 7992 (default 5000 = 5.0 seconds). 7993 7994 usbcore.nousb [USB] Disable the USB subsystem 7995 7996 usbcore.quirks= 7997 [USB] A list of quirk entries to augment the built-in 7998 usb core quirk list. List entries are separated by 7999 commas. Each entry has the form 8000 VendorID:ProductID:Flags. The IDs are 4-digit hex 8001 numbers and Flags is a set of letters. Each letter 8002 will change the built-in quirk; setting it if it is 8003 clear and clearing it if it is set. The letters have 8004 the following meanings: 8005 a = USB_QUIRK_STRING_FETCH_255 (string 8006 descriptors must not be fetched using 8007 a 255-byte read); 8008 b = USB_QUIRK_RESET_RESUME (device can't resume 8009 correctly so reset it instead); 8010 c = USB_QUIRK_NO_SET_INTF (device can't handle 8011 Set-Interface requests); 8012 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't 8013 handle its Configuration or Interface 8014 strings); 8015 e = USB_QUIRK_RESET (device can't be reset 8016 (e.g morph devices), don't use reset); 8017 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has 8018 more interface descriptions than the 8019 bNumInterfaces count, and can't handle 8020 talking to these interfaces); 8021 g = USB_QUIRK_DELAY_INIT (device needs a pause 8022 during initialization, after we read 8023 the device descriptor); 8024 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For 8025 high speed and super speed interrupt 8026 endpoints, the USB 2.0 and USB 3.0 spec 8027 require the interval in microframes (1 8028 microframe = 125 microseconds) to be 8029 calculated as interval = 2 ^ 8030 (bInterval-1). 8031 Devices with this quirk report their 8032 bInterval as the result of this 8033 calculation instead of the exponent 8034 variable used in the calculation); 8035 i = USB_QUIRK_DEVICE_QUALIFIER (device can't 8036 handle device_qualifier descriptor 8037 requests); 8038 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device 8039 generates spurious wakeup, ignore 8040 remote wakeup capability); 8041 k = USB_QUIRK_NO_LPM (device can't handle Link 8042 Power Management); 8043 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL 8044 (Device reports its bInterval as linear 8045 frames instead of the USB 2.0 8046 calculation); 8047 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs 8048 to be disconnected before suspend to 8049 prevent spurious wakeup); 8050 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a 8051 pause after every control message); 8052 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra 8053 delay after resetting its port); 8054 p = USB_QUIRK_SHORT_SET_ADDRESS_REQ_TIMEOUT 8055 (Reduce timeout of the SET_ADDRESS 8056 request from 5000 ms to 500 ms); 8057 Example: quirks=0781:5580:bk,0a5c:5834:gij 8058 8059 usbhid.mousepoll= 8060 [USBHID] The interval which mice are to be polled at. 8061 8062 usbhid.jspoll= 8063 [USBHID] The interval which joysticks are to be polled at. 8064 8065 usbhid.kbpoll= 8066 [USBHID] The interval which keyboards are to be polled at. 8067 8068 usb-storage.delay_use= 8069 [UMS] The delay in seconds before a new device is 8070 scanned for Logical Units (default 1). 8071 Optionally the delay in milliseconds if the value has 8072 suffix with "ms". 8073 Example: delay_use=2567ms 8074 8075 usb-storage.quirks= 8076 [UMS] A list of quirks entries to supplement or 8077 override the built-in unusual_devs list. List 8078 entries are separated by commas. Each entry has 8079 the form VID:PID:Flags where VID and PID are Vendor 8080 and Product ID values (4-digit hex numbers) and 8081 Flags is a set of characters, each corresponding 8082 to a common usb-storage quirk flag as follows: 8083 a = SANE_SENSE (collect more than 18 bytes 8084 of sense data, not on uas); 8085 b = BAD_SENSE (don't collect more than 18 8086 bytes of sense data, not on uas); 8087 c = FIX_CAPACITY (decrease the reported 8088 device capacity by one sector); 8089 d = NO_READ_DISC_INFO (don't use 8090 READ_DISC_INFO command, not on uas); 8091 e = NO_READ_CAPACITY_16 (don't use 8092 READ_CAPACITY_16 command); 8093 f = NO_REPORT_OPCODES (don't use report opcodes 8094 command, uas only); 8095 g = MAX_SECTORS_240 (don't transfer more than 8096 240 sectors at a time, uas only); 8097 h = CAPACITY_HEURISTICS (decrease the 8098 reported device capacity by one 8099 sector if the number is odd); 8100 i = IGNORE_DEVICE (don't bind to this 8101 device); 8102 j = NO_REPORT_LUNS (don't use report luns 8103 command, uas only); 8104 k = NO_SAME (do not use WRITE_SAME, uas only) 8105 l = NOT_LOCKABLE (don't try to lock and 8106 unlock ejectable media, not on uas); 8107 m = MAX_SECTORS_64 (don't transfer more 8108 than 64 sectors = 32 KB at a time, 8109 not on uas); 8110 n = INITIAL_READ10 (force a retry of the 8111 initial READ(10) command, not on uas); 8112 o = CAPACITY_OK (accept the capacity 8113 reported by the device, not on uas); 8114 p = WRITE_CACHE (the device cache is ON 8115 by default, not on uas); 8116 r = IGNORE_RESIDUE (the device reports 8117 bogus residue values, not on uas); 8118 s = SINGLE_LUN (the device has only one 8119 Logical Unit); 8120 t = NO_ATA_1X (don't allow ATA(12) and ATA(16) 8121 commands, uas only); 8122 u = IGNORE_UAS (don't bind to the uas driver); 8123 w = NO_WP_DETECT (don't test whether the 8124 medium is write-protected). 8125 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE 8126 even if the device claims no cache, 8127 not on uas) 8128 Example: quirks=0419:aaf5:rl,0421:0433:rc 8129 8130 user_debug= [KNL,ARM] 8131 Format: <int> 8132 See arch/arm/Kconfig.debug help text. 8133 1 - undefined instruction events 8134 2 - system calls 8135 4 - invalid data aborts 8136 8 - SIGSEGV faults 8137 16 - SIGBUS faults 8138 Example: user_debug=31 8139 8140 vdso= [X86,SH,SPARC] 8141 On X86_32, this is an alias for vdso32=. Otherwise: 8142 8143 vdso=1: enable VDSO (the default) 8144 vdso=0: disable VDSO mapping 8145 8146 vdso32= [X86] Control the 32-bit vDSO 8147 vdso32=1: enable 32-bit VDSO 8148 vdso32=0 or vdso32=2: disable 32-bit VDSO 8149 8150 See the help text for CONFIG_COMPAT_VDSO for more 8151 details. If CONFIG_COMPAT_VDSO is set, the default is 8152 vdso32=0; otherwise, the default is vdso32=1. 8153 8154 For compatibility with older kernels, vdso32=2 is an 8155 alias for vdso32=0. 8156 8157 Try vdso32=0 if you encounter an error that says: 8158 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed! 8159 8160 video= [FB,EARLY] Frame buffer configuration 8161 See Documentation/fb/modedb.rst. 8162 8163 video.brightness_switch_enabled= [ACPI] 8164 Format: [0|1] 8165 If set to 1, on receiving an ACPI notify event 8166 generated by hotkey, video driver will adjust brightness 8167 level and then send out the event to user space through 8168 the allocated input device. If set to 0, video driver 8169 will only send out the event without touching backlight 8170 brightness level. 8171 default: 1 8172 8173 virtio_mmio.device= 8174 [VMMIO] Memory mapped virtio (platform) device. 8175 8176 <size>@<baseaddr>:<irq>[:<id>] 8177 where: 8178 <size> := size (can use standard suffixes 8179 like K, M and G) 8180 <baseaddr> := physical base address 8181 <irq> := interrupt number (as passed to 8182 request_irq()) 8183 <id> := (optional) platform device id 8184 example: 8185 virtio_mmio.device=1K@0x100b0000:48:7 8186 8187 Can be used multiple times for multiple devices. 8188 8189 vga= [BOOT,X86-32] Select a particular video mode 8190 See Documentation/arch/x86/boot.rst and 8191 Documentation/admin-guide/svga.rst. 8192 Use vga=ask for menu. 8193 This is actually a boot loader parameter; the value is 8194 passed to the kernel using a special protocol. 8195 8196 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y. 8197 May slow down system boot speed, especially when 8198 enabled on systems with a large amount of memory. 8199 All options are enabled by default, and this 8200 interface is meant to allow for selectively 8201 enabling or disabling specific virtual memory 8202 debugging features. 8203 8204 Available options are: 8205 P Enable page structure init time poisoning 8206 - Disable all of the above options 8207 8208 vmalloc=nn[KMG] [KNL,BOOT,EARLY] Forces the vmalloc area to have an 8209 exact size of <nn>. This can be used to increase 8210 the minimum size (128MB on x86, arm32 platforms). 8211 It can also be used to decrease the size and leave more room 8212 for directly mapped kernel RAM. Note that this parameter does 8213 not exist on many other platforms (including arm64, alpha, 8214 loongarch, arc, csky, hexagon, microblaze, mips, nios2, openrisc, 8215 parisc, m64k, powerpc, riscv, sh, um, xtensa, s390, sparc). 8216 8217 vmcp_cma=nn[MG] [KNL,S390,EARLY] 8218 Sets the memory size reserved for contiguous memory 8219 allocations for the vmcp device driver. 8220 8221 vmhalt= [KNL,S390] Perform z/VM CP command after system halt. 8222 Format: <command> 8223 8224 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic. 8225 Format: <command> 8226 8227 vmpoff= [KNL,S390] Perform z/VM CP command after power off. 8228 Format: <command> 8229 8230 vmscape= [X86] Controls mitigation for VMscape attacks. 8231 VMscape attacks can leak information from a userspace 8232 hypervisor to a guest via speculative side-channels. 8233 8234 off - disable the mitigation 8235 ibpb - use Indirect Branch Prediction Barrier 8236 (IBPB) mitigation (default) 8237 force - force vulnerability detection even on 8238 unaffected processors 8239 8240 vsyscall= [X86-64,EARLY] 8241 Controls the behavior of vsyscalls (i.e. calls to 8242 fixed addresses of 0xffffffffff600x00 from legacy 8243 code). Most statically-linked binaries and older 8244 versions of glibc use these calls. Because these 8245 functions are at fixed addresses, they make nice 8246 targets for exploits that can control RIP. 8247 8248 emulate Vsyscalls turn into traps and are emulated 8249 reasonably safely. The vsyscall page is 8250 readable. 8251 8252 xonly [default] Vsyscalls turn into traps and are 8253 emulated reasonably safely. The vsyscall 8254 page is not readable. 8255 8256 none Vsyscalls don't work at all. This makes 8257 them quite hard to use for exploits but 8258 might break your system. 8259 8260 vt.color= [VT] Default text color. 8261 Format: 0xYX, X = foreground, Y = background. 8262 Default: 0x07 = light gray on black. 8263 8264 vt.cur_default= [VT] Default cursor shape. 8265 Format: 0xCCBBAA, where AA, BB, and CC are the same as 8266 the parameters of the <Esc>[?A;B;Cc escape sequence; 8267 see vga-softcursor.rst. Default: 2 = underline. 8268 8269 vt.default_blu= [VT] 8270 Format: <blue0>,<blue1>,<blue2>,...,<blue15> 8271 Change the default blue palette of the console. 8272 This is a 16-member array composed of values 8273 ranging from 0-255. 8274 8275 vt.default_grn= [VT] 8276 Format: <green0>,<green1>,<green2>,...,<green15> 8277 Change the default green palette of the console. 8278 This is a 16-member array composed of values 8279 ranging from 0-255. 8280 8281 vt.default_red= [VT] 8282 Format: <red0>,<red1>,<red2>,...,<red15> 8283 Change the default red palette of the console. 8284 This is a 16-member array composed of values 8285 ranging from 0-255. 8286 8287 vt.default_utf8= 8288 [VT] 8289 Format=<0|1> 8290 Set system-wide default UTF-8 mode for all tty's. 8291 Default is 1, i.e. UTF-8 mode is enabled for all 8292 newly opened terminals. 8293 8294 vt.global_cursor_default= 8295 [VT] 8296 Format=<-1|0|1> 8297 Set system-wide default for whether a cursor 8298 is shown on new VTs. Default is -1, 8299 i.e. cursors will be created by default unless 8300 overridden by individual drivers. 0 will hide 8301 cursors, 1 will display them. 8302 8303 vt.italic= [VT] Default color for italic text; 0-15. 8304 Default: 2 = green. 8305 8306 vt.underline= [VT] Default color for underlined text; 0-15. 8307 Default: 3 = cyan. 8308 8309 watchdog timers [HW,WDT] For information on watchdog timers, 8310 see Documentation/watchdog/watchdog-parameters.rst 8311 or other driver-specific files in the 8312 Documentation/watchdog/ directory. 8313 8314 watchdog_thresh= 8315 [KNL] 8316 Set the hard lockup detector stall duration 8317 threshold in seconds. The soft lockup detector 8318 threshold is set to twice the value. A value of 0 8319 disables both lockup detectors. Default is 10 8320 seconds. 8321 8322 workqueue.unbound_cpus= 8323 [KNL,SMP] Specify to constrain one or some CPUs 8324 to use in unbound workqueues. 8325 Format: <cpu-list> 8326 By default, all online CPUs are available for 8327 unbound workqueues. 8328 8329 workqueue.watchdog_thresh= 8330 If CONFIG_WQ_WATCHDOG is configured, workqueue can 8331 warn stall conditions and dump internal state to 8332 help debugging. 0 disables workqueue stall 8333 detection; otherwise, it's the stall threshold 8334 duration in seconds. The default value is 30 and 8335 it can be updated at runtime by writing to the 8336 corresponding sysfs file. 8337 8338 workqueue.panic_on_stall=<uint> 8339 Panic when workqueue stall is detected by 8340 CONFIG_WQ_WATCHDOG. It sets the number times of the 8341 stall to trigger panic. 8342 8343 The default is 0, which disables the panic on stall. 8344 8345 workqueue.cpu_intensive_thresh_us= 8346 Per-cpu work items which run for longer than this 8347 threshold are automatically considered CPU intensive 8348 and excluded from concurrency management to prevent 8349 them from noticeably delaying other per-cpu work 8350 items. Default is 10000 (10ms). 8351 8352 If CONFIG_WQ_CPU_INTENSIVE_REPORT is set, the kernel 8353 will report the work functions which violate this 8354 threshold repeatedly. They are likely good 8355 candidates for using WQ_UNBOUND workqueues instead. 8356 8357 workqueue.cpu_intensive_warning_thresh=<uint> 8358 If CONFIG_WQ_CPU_INTENSIVE_REPORT is set, the kernel 8359 will report the work functions which violate the 8360 intensive_threshold_us repeatedly. In order to prevent 8361 spurious warnings, start printing only after a work 8362 function has violated this threshold number of times. 8363 8364 The default is 4 times. 0 disables the warning. 8365 8366 workqueue.power_efficient 8367 Per-cpu workqueues are generally preferred because 8368 they show better performance thanks to cache 8369 locality; unfortunately, per-cpu workqueues tend to 8370 be more power hungry than unbound workqueues. 8371 8372 Enabling this makes the per-cpu workqueues which 8373 were observed to contribute significantly to power 8374 consumption unbound, leading to measurably lower 8375 power usage at the cost of small performance 8376 overhead. 8377 8378 The default value of this parameter is determined by 8379 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT. 8380 8381 workqueue.default_affinity_scope= 8382 Select the default affinity scope to use for unbound 8383 workqueues. Can be one of "cpu", "smt", "cache", 8384 "numa" and "system". Default is "cache". For more 8385 information, see the Affinity Scopes section in 8386 Documentation/core-api/workqueue.rst. 8387 8388 This can be changed after boot by writing to the 8389 matching /sys/module/workqueue/parameters file. All 8390 workqueues with the "default" affinity scope will be 8391 updated accordingly. 8392 8393 workqueue.debug_force_rr_cpu 8394 Workqueue used to implicitly guarantee that work 8395 items queued without explicit CPU specified are put 8396 on the local CPU. This guarantee is no longer true 8397 and while local CPU is still preferred work items 8398 may be put on foreign CPUs. This debug option 8399 forces round-robin CPU selection to flush out 8400 usages which depend on the now broken guarantee. 8401 When enabled, memory and cache locality will be 8402 impacted. 8403 8404 writecombine= [LOONGARCH,EARLY] Control the MAT (Memory Access 8405 Type) of ioremap_wc(). 8406 8407 on - Enable writecombine, use WUC for ioremap_wc() 8408 off - Disable writecombine, use SUC for ioremap_wc() 8409 8410 x2apic_phys [X86-64,APIC,EARLY] Use x2apic physical mode instead of 8411 default x2apic cluster mode on platforms 8412 supporting x2apic. 8413 8414 xen_512gb_limit [KNL,X86-64,XEN] 8415 Restricts the kernel running paravirtualized under Xen 8416 to use only up to 512 GB of RAM. The reason to do so is 8417 crash analysis tools and Xen tools for doing domain 8418 save/restore/migration must be enabled to handle larger 8419 domains. 8420 8421 xen_emul_unplug= [HW,X86,XEN,EARLY] 8422 Unplug Xen emulated devices 8423 Format: [unplug0,][unplug1] 8424 ide-disks -- unplug primary master IDE devices 8425 aux-ide-disks -- unplug non-primary-master IDE devices 8426 nics -- unplug network devices 8427 all -- unplug all emulated devices (NICs and IDE disks) 8428 unnecessary -- unplugging emulated devices is 8429 unnecessary even if the host did not respond to 8430 the unplug protocol 8431 never -- do not unplug even if version check succeeds 8432 8433 xen_legacy_crash [X86,XEN,EARLY] 8434 Crash from Xen panic notifier, without executing late 8435 panic() code such as dumping handler. 8436 8437 xen_mc_debug [X86,XEN,EARLY] 8438 Enable multicall debugging when running as a Xen PV guest. 8439 Enabling this feature will reduce performance a little 8440 bit, so it should only be enabled for obtaining extended 8441 debug data in case of multicall errors. 8442 8443 xen_msr_safe= [X86,XEN,EARLY] 8444 Format: <bool> 8445 Select whether to always use non-faulting (safe) MSR 8446 access functions when running as Xen PV guest. The 8447 default value is controlled by CONFIG_XEN_PV_MSR_SAFE. 8448 8449 xen_nopv [X86] 8450 Disables the PV optimizations forcing the HVM guest to 8451 run as generic HVM guest with no PV drivers. 8452 This option is obsoleted by the "nopv" option, which 8453 has equivalent effect for XEN platform. 8454 8455 xen_no_vector_callback 8456 [KNL,X86,XEN,EARLY] Disable the vector callback for Xen 8457 event channel interrupts. 8458 8459 xen_scrub_pages= [XEN] 8460 Boolean option to control scrubbing pages before giving them back 8461 to Xen, for use by other domains. Can be also changed at runtime 8462 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages. 8463 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT. 8464 8465 xen_timer_slop= [X86-64,XEN,EARLY] 8466 Set the timer slop (in nanoseconds) for the virtual Xen 8467 timers (default is 100000). This adjusts the minimum 8468 delta of virtualized Xen timers, where lower values 8469 improve timer resolution at the expense of processing 8470 more timer interrupts. 8471 8472 xen.balloon_boot_timeout= [XEN] 8473 The time (in seconds) to wait before giving up to boot 8474 in case initial ballooning fails to free enough memory. 8475 Applies only when running as HVM or PVH guest and 8476 started with less memory configured than allowed at 8477 max. Default is 180. 8478 8479 xen.event_eoi_delay= [XEN] 8480 How long to delay EOI handling in case of event 8481 storms (jiffies). Default is 10. 8482 8483 xen.event_loop_timeout= [XEN] 8484 After which time (jiffies) the event handling loop 8485 should start to delay EOI handling. Default is 2. 8486 8487 xen.fifo_events= [XEN] 8488 Boolean parameter to disable using fifo event handling 8489 even if available. Normally fifo event handling is 8490 preferred over the 2-level event handling, as it is 8491 fairer and the number of possible event channels is 8492 much higher. Default is on (use fifo events). 8493 8494 xirc2ps_cs= [NET,PCMCIA] 8495 Format: 8496 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]] 8497 8498 xive= [PPC] 8499 By default on POWER9 and above, the kernel will 8500 natively use the XIVE interrupt controller. This option 8501 allows the fallback firmware mode to be used: 8502 8503 off Fallback to firmware control of XIVE interrupt 8504 controller on both pseries and powernv 8505 platforms. Only useful on POWER9 and above. 8506 8507 xive.store-eoi=off [PPC] 8508 By default on POWER10 and above, the kernel will use 8509 stores for EOI handling when the XIVE interrupt mode 8510 is active. This option allows the XIVE driver to use 8511 loads instead, as on POWER9. 8512 8513 xhci-hcd.quirks [USB,KNL] 8514 A hex value specifying bitmask with supplemental xhci 8515 host controller quirks. Meaning of each bit can be 8516 consulted in header drivers/usb/host/xhci.h. 8517 8518 xmon [PPC,EARLY] 8519 Format: { early | on | rw | ro | off } 8520 Controls if xmon debugger is enabled. Default is off. 8521 Passing only "xmon" is equivalent to "xmon=early". 8522 early Call xmon as early as possible on boot; xmon 8523 debugger is called from setup_arch(). 8524 on xmon debugger hooks will be installed so xmon 8525 is only called on a kernel crash. Default mode, 8526 i.e. either "ro" or "rw" mode, is controlled 8527 with CONFIG_XMON_DEFAULT_RO_MODE. 8528 rw xmon debugger hooks will be installed so xmon 8529 is called only on a kernel crash, mode is write, 8530 meaning SPR registers, memory and, other data 8531 can be written using xmon commands. 8532 ro same as "rw" option above but SPR registers, 8533 memory, and other data can't be written using 8534 xmon commands. 8535 off xmon is disabled. 8536