1 acpi= [HW,ACPI,X86,ARM64,RISCV64] 2 Advanced Configuration and Power Interface 3 Format: { force | on | off | strict | noirq | rsdt | 4 copy_dsdt } 5 force -- enable ACPI if default was off 6 on -- enable ACPI but allow fallback to DT [arm64,riscv64] 7 off -- disable ACPI if default was on 8 noirq -- do not use ACPI for IRQ routing 9 strict -- Be less tolerant of platforms that are not 10 strictly ACPI specification compliant. 11 rsdt -- prefer RSDT over (default) XSDT 12 copy_dsdt -- copy DSDT to memory 13 For ARM64 and RISCV64, ONLY "acpi=off", "acpi=on" or 14 "acpi=force" are available 15 16 See also Documentation/power/runtime_pm.rst, pci=noacpi 17 18 acpi_apic_instance= [ACPI, IOAPIC] 19 Format: <int> 20 2: use 2nd APIC table, if available 21 1,0: use 1st APIC table 22 default: 0 23 24 acpi_backlight= [HW,ACPI] 25 { vendor | video | native | none } 26 If set to vendor, prefer vendor-specific driver 27 (e.g. thinkpad_acpi, sony_acpi, etc.) instead 28 of the ACPI video.ko driver. 29 If set to video, use the ACPI video.ko driver. 30 If set to native, use the device's native backlight mode. 31 If set to none, disable the ACPI backlight interface. 32 33 acpi_force_32bit_fadt_addr 34 force FADT to use 32 bit addresses rather than the 35 64 bit X_* addresses. Some firmware have broken 64 36 bit addresses for force ACPI ignore these and use 37 the older legacy 32 bit addresses. 38 39 acpica_no_return_repair [HW, ACPI] 40 Disable AML predefined validation mechanism 41 This mechanism can repair the evaluation result to make 42 the return objects more ACPI specification compliant. 43 This option is useful for developers to identify the 44 root cause of an AML interpreter issue when the issue 45 has something to do with the repair mechanism. 46 47 acpi.debug_layer= [HW,ACPI,ACPI_DEBUG] 48 acpi.debug_level= [HW,ACPI,ACPI_DEBUG] 49 Format: <int> 50 CONFIG_ACPI_DEBUG must be enabled to produce any ACPI 51 debug output. Bits in debug_layer correspond to a 52 _COMPONENT in an ACPI source file, e.g., 53 #define _COMPONENT ACPI_EVENTS 54 Bits in debug_level correspond to a level in 55 ACPI_DEBUG_PRINT statements, e.g., 56 ACPI_DEBUG_PRINT((ACPI_DB_INFO, ... 57 The debug_level mask defaults to "info". See 58 Documentation/firmware-guide/acpi/debug.rst for more information about 59 debug layers and levels. 60 61 Enable processor driver info messages: 62 acpi.debug_layer=0x20000000 63 Enable AML "Debug" output, i.e., stores to the Debug 64 object while interpreting AML: 65 acpi.debug_layer=0xffffffff acpi.debug_level=0x2 66 Enable all messages related to ACPI hardware: 67 acpi.debug_layer=0x2 acpi.debug_level=0xffffffff 68 69 Some values produce so much output that the system is 70 unusable. The "log_buf_len" parameter may be useful 71 if you need to capture more output. 72 73 acpi_enforce_resources= [ACPI] 74 { strict | lax | no } 75 Check for resource conflicts between native drivers 76 and ACPI OperationRegions (SystemIO and SystemMemory 77 only). IO ports and memory declared in ACPI might be 78 used by the ACPI subsystem in arbitrary AML code and 79 can interfere with legacy drivers. 80 strict (default): access to resources claimed by ACPI 81 is denied; legacy drivers trying to access reserved 82 resources will fail to bind to device using them. 83 lax: access to resources claimed by ACPI is allowed; 84 legacy drivers trying to access reserved resources 85 will bind successfully but a warning message is logged. 86 no: ACPI OperationRegions are not marked as reserved, 87 no further checks are performed. 88 89 acpi_force_table_verification [HW,ACPI] 90 Enable table checksum verification during early stage. 91 By default, this is disabled due to x86 early mapping 92 size limitation. 93 94 acpi_irq_balance [HW,ACPI] 95 ACPI will balance active IRQs 96 default in APIC mode 97 98 acpi_irq_nobalance [HW,ACPI] 99 ACPI will not move active IRQs (default) 100 default in PIC mode 101 102 acpi_irq_isa= [HW,ACPI] If irq_balance, mark listed IRQs used by ISA 103 Format: <irq>,<irq>... 104 105 acpi_irq_pci= [HW,ACPI] If irq_balance, clear listed IRQs for 106 use by PCI 107 Format: <irq>,<irq>... 108 109 acpi_mask_gpe= [HW,ACPI] 110 Due to the existence of _Lxx/_Exx, some GPEs triggered 111 by unsupported hardware/firmware features can result in 112 GPE floodings that cannot be automatically disabled by 113 the GPE dispatcher. 114 This facility can be used to prevent such uncontrolled 115 GPE floodings. 116 Format: <byte> or <bitmap-list> 117 118 acpi_no_auto_serialize [HW,ACPI] 119 Disable auto-serialization of AML methods 120 AML control methods that contain the opcodes to create 121 named objects will be marked as "Serialized" by the 122 auto-serialization feature. 123 This feature is enabled by default. 124 This option allows to turn off the feature. 125 126 acpi_no_memhotplug [ACPI] Disable memory hotplug. Useful for kdump 127 kernels. 128 129 acpi_no_static_ssdt [HW,ACPI] 130 Disable installation of static SSDTs at early boot time 131 By default, SSDTs contained in the RSDT/XSDT will be 132 installed automatically and they will appear under 133 /sys/firmware/acpi/tables. 134 This option turns off this feature. 135 Note that specifying this option does not affect 136 dynamic table installation which will install SSDT 137 tables to /sys/firmware/acpi/tables/dynamic. 138 139 acpi_no_watchdog [HW,ACPI,WDT] 140 Ignore the ACPI-based watchdog interface (WDAT) and let 141 a native driver control the watchdog device instead. 142 143 acpi_rsdp= [ACPI,EFI,KEXEC] 144 Pass the RSDP address to the kernel, mostly used 145 on machines running EFI runtime service to boot the 146 second kernel for kdump. 147 148 acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS 149 Format: To spoof as Windows 98: ="Microsoft Windows" 150 151 acpi_rev_override [ACPI] Override the _REV object to return 5 (instead 152 of 2 which is mandated by ACPI 6) as the supported ACPI 153 specification revision (when using this switch, it may 154 be necessary to carry out a cold reboot _twice_ in a 155 row to make it take effect on the platform firmware). 156 157 acpi_osi= [HW,ACPI] Modify list of supported OS interface strings 158 acpi_osi="string1" # add string1 159 acpi_osi="!string2" # remove string2 160 acpi_osi=!* # remove all strings 161 acpi_osi=! # disable all built-in OS vendor 162 strings 163 acpi_osi=!! # enable all built-in OS vendor 164 strings 165 acpi_osi= # disable all strings 166 167 'acpi_osi=!' can be used in combination with single or 168 multiple 'acpi_osi="string1"' to support specific OS 169 vendor string(s). Note that such command can only 170 affect the default state of the OS vendor strings, thus 171 it cannot affect the default state of the feature group 172 strings and the current state of the OS vendor strings, 173 specifying it multiple times through kernel command line 174 is meaningless. This command is useful when one do not 175 care about the state of the feature group strings which 176 should be controlled by the OSPM. 177 Examples: 178 1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent 179 to 'acpi_osi="Windows 2000" acpi_osi=!', they all 180 can make '_OSI("Windows 2000")' TRUE. 181 182 'acpi_osi=' cannot be used in combination with other 183 'acpi_osi=' command lines, the _OSI method will not 184 exist in the ACPI namespace. NOTE that such command can 185 only affect the _OSI support state, thus specifying it 186 multiple times through kernel command line is also 187 meaningless. 188 Examples: 189 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)' 190 FALSE. 191 192 'acpi_osi=!*' can be used in combination with single or 193 multiple 'acpi_osi="string1"' to support specific 194 string(s). Note that such command can affect the 195 current state of both the OS vendor strings and the 196 feature group strings, thus specifying it multiple times 197 through kernel command line is meaningful. But it may 198 still not able to affect the final state of a string if 199 there are quirks related to this string. This command 200 is useful when one want to control the state of the 201 feature group strings to debug BIOS issues related to 202 the OSPM features. 203 Examples: 204 1. 'acpi_osi="Module Device" acpi_osi=!*' can make 205 '_OSI("Module Device")' FALSE. 206 2. 'acpi_osi=!* acpi_osi="Module Device"' can make 207 '_OSI("Module Device")' TRUE. 208 3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is 209 equivalent to 210 'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"' 211 and 212 'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!', 213 they all will make '_OSI("Windows 2000")' TRUE. 214 215 acpi_pm_good [X86] 216 Override the pmtimer bug detection: force the kernel 217 to assume that this machine's pmtimer latches its value 218 and always returns good values. 219 220 acpi_sci= [HW,ACPI] ACPI System Control Interrupt trigger mode 221 Format: { level | edge | high | low } 222 223 acpi_skip_timer_override [HW,ACPI] 224 Recognize and ignore IRQ0/pin2 Interrupt Override. 225 For broken nForce2 BIOS resulting in XT-PIC timer. 226 227 acpi_sleep= [HW,ACPI] Sleep options 228 Format: { s3_bios, s3_mode, s3_beep, s4_hwsig, 229 s4_nohwsig, old_ordering, nonvs, 230 sci_force_enable, nobl } 231 See Documentation/power/video.rst for information on 232 s3_bios and s3_mode. 233 s3_beep is for debugging; it makes the PC's speaker beep 234 as soon as the kernel's real-mode entry point is called. 235 s4_hwsig causes the kernel to check the ACPI hardware 236 signature during resume from hibernation, and gracefully 237 refuse to resume if it has changed. This complies with 238 the ACPI specification but not with reality, since 239 Windows does not do this and many laptops do change it 240 on docking. So the default behaviour is to allow resume 241 and simply warn when the signature changes, unless the 242 s4_hwsig option is enabled. 243 s4_nohwsig prevents ACPI hardware signature from being 244 used (or even warned about) during resume. 245 old_ordering causes the ACPI 1.0 ordering of the _PTS 246 control method, with respect to putting devices into 247 low power states, to be enforced (the ACPI 2.0 ordering 248 of _PTS is used by default). 249 nonvs prevents the kernel from saving/restoring the 250 ACPI NVS memory during suspend/hibernation and resume. 251 sci_force_enable causes the kernel to set SCI_EN directly 252 on resume from S1/S3 (which is against the ACPI spec, 253 but some broken systems don't work without it). 254 nobl causes the internal blacklist of systems known to 255 behave incorrectly in some ways with respect to system 256 suspend and resume to be ignored (use wisely). 257 258 acpi_use_timer_override [HW,ACPI] 259 Use timer override. For some broken Nvidia NF5 boards 260 that require a timer override, but don't have HPET 261 262 add_efi_memmap [EFI; X86] Include EFI memory map in 263 kernel's map of available physical RAM. 264 265 agp= [AGP] 266 { off | try_unsupported } 267 off: disable AGP support 268 try_unsupported: try to drive unsupported chipsets 269 (may crash computer or cause data corruption) 270 271 ALSA [HW,ALSA] 272 See Documentation/sound/alsa-configuration.rst 273 274 alignment= [KNL,ARM] 275 Allow the default userspace alignment fault handler 276 behaviour to be specified. Bit 0 enables warnings, 277 bit 1 enables fixups, and bit 2 sends a segfault. 278 279 align_va_addr= [X86-64] 280 Align virtual addresses by clearing slice [14:12] when 281 allocating a VMA at process creation time. This option 282 gives you up to 3% performance improvement on AMD F15h 283 machines (where it is enabled by default) for a 284 CPU-intensive style benchmark, and it can vary highly in 285 a microbenchmark depending on workload and compiler. 286 287 32: only for 32-bit processes 288 64: only for 64-bit processes 289 on: enable for both 32- and 64-bit processes 290 off: disable for both 32- and 64-bit processes 291 292 alloc_snapshot [FTRACE] 293 Allocate the ftrace snapshot buffer on boot up when the 294 main buffer is allocated. This is handy if debugging 295 and you need to use tracing_snapshot() on boot up, and 296 do not want to use tracing_snapshot_alloc() as it needs 297 to be done where GFP_KERNEL allocations are allowed. 298 299 allow_mismatched_32bit_el0 [ARM64] 300 Allow execve() of 32-bit applications and setting of the 301 PER_LINUX32 personality on systems where only a strict 302 subset of the CPUs support 32-bit EL0. When this 303 parameter is present, the set of CPUs supporting 32-bit 304 EL0 is indicated by /sys/devices/system/cpu/aarch32_el0 305 and hot-unplug operations may be restricted. 306 307 See Documentation/arch/arm64/asymmetric-32bit.rst for more 308 information. 309 310 amd_iommu= [HW,X86-64] 311 Pass parameters to the AMD IOMMU driver in the system. 312 Possible values are: 313 fullflush - Deprecated, equivalent to iommu.strict=1 314 off - do not initialize any AMD IOMMU found in 315 the system 316 force_isolation - Force device isolation for all 317 devices. The IOMMU driver is not 318 allowed anymore to lift isolation 319 requirements as needed. This option 320 does not override iommu=pt 321 force_enable - Force enable the IOMMU on platforms known 322 to be buggy with IOMMU enabled. Use this 323 option with care. 324 pgtbl_v1 - Use v1 page table for DMA-API (Default). 325 pgtbl_v2 - Use v2 page table for DMA-API. 326 irtcachedis - Disable Interrupt Remapping Table (IRT) caching. 327 328 amd_iommu_dump= [HW,X86-64] 329 Enable AMD IOMMU driver option to dump the ACPI table 330 for AMD IOMMU. With this option enabled, AMD IOMMU 331 driver will print ACPI tables for AMD IOMMU during 332 IOMMU initialization. 333 334 amd_iommu_intr= [HW,X86-64] 335 Specifies one of the following AMD IOMMU interrupt 336 remapping modes: 337 legacy - Use legacy interrupt remapping mode. 338 vapic - Use virtual APIC mode, which allows IOMMU 339 to inject interrupts directly into guest. 340 This mode requires kvm-amd.avic=1. 341 (Default when IOMMU HW support is present.) 342 343 amd_pstate= [X86] 344 disable 345 Do not enable amd_pstate as the default 346 scaling driver for the supported processors 347 passive 348 Use amd_pstate with passive mode as a scaling driver. 349 In this mode autonomous selection is disabled. 350 Driver requests a desired performance level and platform 351 tries to match the same performance level if it is 352 satisfied by guaranteed performance level. 353 active 354 Use amd_pstate_epp driver instance as the scaling driver, 355 driver provides a hint to the hardware if software wants 356 to bias toward performance (0x0) or energy efficiency (0xff) 357 to the CPPC firmware. then CPPC power algorithm will 358 calculate the runtime workload and adjust the realtime cores 359 frequency. 360 guided 361 Activate guided autonomous mode. Driver requests minimum and 362 maximum performance level and the platform autonomously 363 selects a performance level in this range and appropriate 364 to the current workload. 365 366 amijoy.map= [HW,JOY] Amiga joystick support 367 Map of devices attached to JOY0DAT and JOY1DAT 368 Format: <a>,<b> 369 See also Documentation/input/joydev/joystick.rst 370 371 analog.map= [HW,JOY] Analog joystick and gamepad support 372 Specifies type or capabilities of an analog joystick 373 connected to one of 16 gameports 374 Format: <type1>,<type2>,..<type16> 375 376 apc= [HW,SPARC] 377 Power management functions (SPARCstation-4/5 + deriv.) 378 Format: noidle 379 Disable APC CPU standby support. SPARCstation-Fox does 380 not play well with APC CPU idle - disable it if you have 381 APC and your system crashes randomly. 382 383 apic= [APIC,X86] Advanced Programmable Interrupt Controller 384 Change the output verbosity while booting 385 Format: { quiet (default) | verbose | debug } 386 Change the amount of debugging information output 387 when initialising the APIC and IO-APIC components. 388 For X86-32, this can also be used to specify an APIC 389 driver name. 390 Format: apic=driver_name 391 Examples: apic=bigsmp 392 393 apic_extnmi= [APIC,X86] External NMI delivery setting 394 Format: { bsp (default) | all | none } 395 bsp: External NMI is delivered only to CPU 0 396 all: External NMIs are broadcast to all CPUs as a 397 backup of CPU 0 398 none: External NMI is masked for all CPUs. This is 399 useful so that a dump capture kernel won't be 400 shot down by NMI 401 402 autoconf= [IPV6] 403 See Documentation/networking/ipv6.rst. 404 405 apm= [APM] Advanced Power Management 406 See header of arch/x86/kernel/apm_32.c. 407 408 apparmor= [APPARMOR] Disable or enable AppArmor at boot time 409 Format: { "0" | "1" } 410 See security/apparmor/Kconfig help text 411 0 -- disable. 412 1 -- enable. 413 Default value is set via kernel config option. 414 415 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards 416 Format: <io>,<irq>,<nodeID> 417 418 arm64.nobti [ARM64] Unconditionally disable Branch Target 419 Identification support 420 421 arm64.nomops [ARM64] Unconditionally disable Memory Copy and Memory 422 Set instructions support 423 424 arm64.nomte [ARM64] Unconditionally disable Memory Tagging Extension 425 support 426 427 arm64.nopauth [ARM64] Unconditionally disable Pointer Authentication 428 support 429 430 arm64.nosme [ARM64] Unconditionally disable Scalable Matrix 431 Extension support 432 433 arm64.nosve [ARM64] Unconditionally disable Scalable Vector 434 Extension support 435 436 ataflop= [HW,M68k] 437 438 atarimouse= [HW,MOUSE] Atari Mouse 439 440 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess, 441 EzKey and similar keyboards 442 443 atkbd.reset= [HW] Reset keyboard during initialization 444 445 atkbd.set= [HW] Select keyboard code set 446 Format: <int> (2 = AT (default), 3 = PS/2) 447 448 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar 449 keyboards 450 451 atkbd.softraw= [HW] Choose between synthetic and real raw mode 452 Format: <bool> (0 = real, 1 = synthetic (default)) 453 454 atkbd.softrepeat= [HW] 455 Use software keyboard repeat 456 457 audit= [KNL] Enable the audit sub-system 458 Format: { "0" | "1" | "off" | "on" } 459 0 | off - kernel audit is disabled and can not be 460 enabled until the next reboot 461 unset - kernel audit is initialized but disabled and 462 will be fully enabled by the userspace auditd. 463 1 | on - kernel audit is initialized and partially 464 enabled, storing at most audit_backlog_limit 465 messages in RAM until it is fully enabled by the 466 userspace auditd. 467 Default: unset 468 469 audit_backlog_limit= [KNL] Set the audit queue size limit. 470 Format: <int> (must be >=0) 471 Default: 64 472 473 bau= [X86_UV] Enable the BAU on SGI UV. The default 474 behavior is to disable the BAU (i.e. bau=0). 475 Format: { "0" | "1" } 476 0 - Disable the BAU. 477 1 - Enable the BAU. 478 unset - Disable the BAU. 479 480 baycom_epp= [HW,AX25] 481 Format: <io>,<mode> 482 483 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem 484 Format: <io>,<mode> 485 See header of drivers/net/hamradio/baycom_par.c. 486 487 baycom_ser_fdx= [HW,AX25] 488 BayCom Serial Port AX.25 Modem (Full Duplex Mode) 489 Format: <io>,<irq>,<mode>[,<baud>] 490 See header of drivers/net/hamradio/baycom_ser_fdx.c. 491 492 baycom_ser_hdx= [HW,AX25] 493 BayCom Serial Port AX.25 Modem (Half Duplex Mode) 494 Format: <io>,<irq>,<mode> 495 See header of drivers/net/hamradio/baycom_ser_hdx.c. 496 497 bert_disable [ACPI] 498 Disable BERT OS support on buggy BIOSes. 499 500 bgrt_disable [ACPI][X86] 501 Disable BGRT to avoid flickering OEM logo. 502 503 blkdevparts= Manual partition parsing of block device(s) for 504 embedded devices based on command line input. 505 See Documentation/block/cmdline-partition.rst 506 507 boot_delay= Milliseconds to delay each printk during boot. 508 Only works if CONFIG_BOOT_PRINTK_DELAY is enabled, 509 and you may also have to specify "lpj=". Boot_delay 510 values larger than 10 seconds (10000) are assumed 511 erroneous and ignored. 512 Format: integer 513 514 bootconfig [KNL] 515 Extended command line options can be added to an initrd 516 and this will cause the kernel to look for it. 517 518 See Documentation/admin-guide/bootconfig.rst 519 520 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards) 521 bttv.radio= Most important insmod options are available as 522 kernel args too. 523 bttv.pll= See Documentation/admin-guide/media/bttv.rst 524 bttv.tuner= 525 526 bulk_remove=off [PPC] This parameter disables the use of the pSeries 527 firmware feature for flushing multiple hpte entries 528 at a time. 529 530 c101= [NET] Moxa C101 synchronous serial card 531 532 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection. 533 Sometimes CPU hardware bugs make them report the cache 534 size incorrectly. The kernel will attempt work arounds 535 to fix known problems, but for some CPUs it is not 536 possible to determine what the correct size should be. 537 This option provides an override for these situations. 538 539 carrier_timeout= 540 [NET] Specifies amount of time (in seconds) that 541 the kernel should wait for a network carrier. By default 542 it waits 120 seconds. 543 544 ca_keys= [KEYS] This parameter identifies a specific key(s) on 545 the system trusted keyring to be used for certificate 546 trust validation. 547 format: { id:<keyid> | builtin } 548 549 cca= [MIPS] Override the kernel pages' cache coherency 550 algorithm. Accepted values range from 0 to 7 551 inclusive. See arch/mips/include/asm/pgtable-bits.h 552 for platform specific values (SB1, Loongson3 and 553 others). 554 555 ccw_timeout_log [S390] 556 See Documentation/arch/s390/common_io.rst for details. 557 558 cgroup_disable= [KNL] Disable a particular controller or optional feature 559 Format: {name of the controller(s) or feature(s) to disable} 560 The effects of cgroup_disable=foo are: 561 - foo isn't auto-mounted if you mount all cgroups in 562 a single hierarchy 563 - foo isn't visible as an individually mountable 564 subsystem 565 - if foo is an optional feature then the feature is 566 disabled and corresponding cgroup files are not 567 created 568 {Currently only "memory" controller deal with this and 569 cut the overhead, others just disable the usage. So 570 only cgroup_disable=memory is actually worthy} 571 Specifying "pressure" disables per-cgroup pressure 572 stall information accounting feature 573 574 cgroup_no_v1= [KNL] Disable cgroup controllers and named hierarchies in v1 575 Format: { { controller | "all" | "named" } 576 [,{ controller | "all" | "named" }...] } 577 Like cgroup_disable, but only applies to cgroup v1; 578 the blacklisted controllers remain available in cgroup2. 579 "all" blacklists all controllers and "named" disables 580 named mounts. Specifying both "all" and "named" disables 581 all v1 hierarchies. 582 583 cgroup.memory= [KNL] Pass options to the cgroup memory controller. 584 Format: <string> 585 nosocket -- Disable socket memory accounting. 586 nokmem -- Disable kernel memory accounting. 587 nobpf -- Disable BPF memory accounting. 588 589 checkreqprot= [SELINUX] Set initial checkreqprot flag value. 590 Format: { "0" | "1" } 591 See security/selinux/Kconfig help text. 592 0 -- check protection applied by kernel (includes 593 any implied execute protection). 594 1 -- check protection requested by application. 595 Default value is set via a kernel config option. 596 Value can be changed at runtime via 597 /sys/fs/selinux/checkreqprot. 598 Setting checkreqprot to 1 is deprecated. 599 600 cio_ignore= [S390] 601 See Documentation/arch/s390/common_io.rst for details. 602 603 clearcpuid=X[,X...] [X86] 604 Disable CPUID feature X for the kernel. See 605 arch/x86/include/asm/cpufeatures.h for the valid bit 606 numbers X. Note the Linux-specific bits are not necessarily 607 stable over kernel options, but the vendor-specific 608 ones should be. 609 X can also be a string as appearing in the flags: line 610 in /proc/cpuinfo which does not have the above 611 instability issue. However, not all features have names 612 in /proc/cpuinfo. 613 Note that using this option will taint your kernel. 614 Also note that user programs calling CPUID directly 615 or using the feature without checking anything 616 will still see it. This just prevents it from 617 being used by the kernel or shown in /proc/cpuinfo. 618 Also note the kernel might malfunction if you disable 619 some critical bits. 620 621 clk_ignore_unused 622 [CLK] 623 Prevents the clock framework from automatically gating 624 clocks that have not been explicitly enabled by a Linux 625 device driver but are enabled in hardware at reset or 626 by the bootloader/firmware. Note that this does not 627 force such clocks to be always-on nor does it reserve 628 those clocks in any way. This parameter is useful for 629 debug and development, but should not be needed on a 630 platform with proper driver support. For more 631 information, see Documentation/driver-api/clk.rst. 632 633 clock= [BUGS=X86-32, HW] gettimeofday clocksource override. 634 [Deprecated] 635 Forces specified clocksource (if available) to be used 636 when calculating gettimeofday(). If specified 637 clocksource is not available, it defaults to PIT. 638 Format: { pit | tsc | cyclone | pmtmr } 639 640 clocksource= Override the default clocksource 641 Format: <string> 642 Override the default clocksource and use the clocksource 643 with the name specified. 644 Some clocksource names to choose from, depending on 645 the platform: 646 [all] jiffies (this is the base, fallback clocksource) 647 [ACPI] acpi_pm 648 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2, 649 pxa_timer,timer3,32k_counter,timer0_1 650 [X86-32] pit,hpet,tsc; 651 scx200_hrt on Geode; cyclone on IBM x440 652 [MIPS] MIPS 653 [PARISC] cr16 654 [S390] tod 655 [SH] SuperH 656 [SPARC64] tick 657 [X86-64] hpet,tsc 658 659 clocksource.arm_arch_timer.evtstrm= 660 [ARM,ARM64] 661 Format: <bool> 662 Enable/disable the eventstream feature of the ARM 663 architected timer so that code using WFE-based polling 664 loops can be debugged more effectively on production 665 systems. 666 667 clocksource.verify_n_cpus= [KNL] 668 Limit the number of CPUs checked for clocksources 669 marked with CLOCK_SOURCE_VERIFY_PERCPU that 670 are marked unstable due to excessive skew. 671 A negative value says to check all CPUs, while 672 zero says not to check any. Values larger than 673 nr_cpu_ids are silently truncated to nr_cpu_ids. 674 The actual CPUs are chosen randomly, with 675 no replacement if the same CPU is chosen twice. 676 677 clocksource-wdtest.holdoff= [KNL] 678 Set the time in seconds that the clocksource 679 watchdog test waits before commencing its tests. 680 Defaults to zero when built as a module and to 681 10 seconds when built into the kernel. 682 683 cma=nn[MG]@[start[MG][-end[MG]]] 684 [KNL,CMA] 685 Sets the size of kernel global memory area for 686 contiguous memory allocations and optionally the 687 placement constraint by the physical address range of 688 memory allocations. A value of 0 disables CMA 689 altogether. For more information, see 690 kernel/dma/contiguous.c 691 692 cma_pernuma=nn[MG] 693 [KNL,CMA] 694 Sets the size of kernel per-numa memory area for 695 contiguous memory allocations. A value of 0 disables 696 per-numa CMA altogether. And If this option is not 697 specified, the default value is 0. 698 With per-numa CMA enabled, DMA users on node nid will 699 first try to allocate buffer from the pernuma area 700 which is located in node nid, if the allocation fails, 701 they will fallback to the global default memory area. 702 703 numa_cma=<node>:nn[MG][,<node>:nn[MG]] 704 [KNL,CMA] 705 Sets the size of kernel numa memory area for 706 contiguous memory allocations. It will reserve CMA 707 area for the specified node. 708 709 With numa CMA enabled, DMA users on node nid will 710 first try to allocate buffer from the numa area 711 which is located in node nid, if the allocation fails, 712 they will fallback to the global default memory area. 713 714 cmo_free_hint= [PPC] Format: { yes | no } 715 Specify whether pages are marked as being inactive 716 when they are freed. This is used in CMO environments 717 to determine OS memory pressure for page stealing by 718 a hypervisor. 719 Default: yes 720 721 coherent_pool=nn[KMG] [ARM,KNL] 722 Sets the size of memory pool for coherent, atomic dma 723 allocations, by default set to 256K. 724 725 com20020= [HW,NET] ARCnet - COM20020 chipset 726 Format: 727 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]] 728 729 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers) 730 Format: <io>[,<irq>] 731 732 com90xx= [HW,NET] 733 ARCnet - COM90xx chipset (memory-mapped buffers) 734 Format: <io>[,<irq>[,<memstart>]] 735 736 condev= [HW,S390] console device 737 conmode= 738 739 con3215_drop= [S390] 3215 console drop mode. 740 Format: y|n|Y|N|1|0 741 When set to true, drop data on the 3215 console when 742 the console buffer is full. In this case the 743 operator using a 3270 terminal emulator (for example 744 x3270) does not have to enter the clear key for the 745 console output to advance and the kernel to continue. 746 This leads to a much faster boot time when a 3270 747 terminal emulator is active. If no 3270 terminal 748 emulator is used, this parameter has no effect. 749 750 console= [KNL] Output console device and options. 751 752 tty<n> Use the virtual console device <n>. 753 754 ttyS<n>[,options] 755 ttyUSB0[,options] 756 Use the specified serial port. The options are of 757 the form "bbbbpnf", where "bbbb" is the baud rate, 758 "p" is parity ("n", "o", or "e"), "n" is number of 759 bits, and "f" is flow control ("r" for RTS or 760 omit it). Default is "9600n8". 761 762 See Documentation/admin-guide/serial-console.rst for more 763 information. See 764 Documentation/networking/netconsole.rst for an 765 alternative. 766 767 uart[8250],io,<addr>[,options] 768 uart[8250],mmio,<addr>[,options] 769 uart[8250],mmio16,<addr>[,options] 770 uart[8250],mmio32,<addr>[,options] 771 uart[8250],0x<addr>[,options] 772 Start an early, polled-mode console on the 8250/16550 773 UART at the specified I/O port or MMIO address, 774 switching to the matching ttyS device later. 775 MMIO inter-register address stride is either 8-bit 776 (mmio), 16-bit (mmio16), or 32-bit (mmio32). 777 If none of [io|mmio|mmio16|mmio32], <addr> is assumed 778 to be equivalent to 'mmio'. 'options' are specified in 779 the same format described for ttyS above; if unspecified, 780 the h/w is not re-initialized. 781 782 hvc<n> Use the hypervisor console device <n>. This is for 783 both Xen and PowerPC hypervisors. 784 785 { null | "" } 786 Use to disable console output, i.e., to have kernel 787 console messages discarded. 788 This must be the only console= parameter used on the 789 kernel command line. 790 791 If the device connected to the port is not a TTY but a braille 792 device, prepend "brl," before the device type, for instance 793 console=brl,ttyS0 794 For now, only VisioBraille is supported. 795 796 console_msg_format= 797 [KNL] Change console messages format 798 default 799 By default we print messages on consoles in 800 "[time stamp] text\n" format (time stamp may not be 801 printed, depending on CONFIG_PRINTK_TIME or 802 `printk_time' param). 803 syslog 804 Switch to syslog format: "<%u>[time stamp] text\n" 805 IOW, each message will have a facility and loglevel 806 prefix. The format is similar to one used by syslog() 807 syscall, or to executing "dmesg -S --raw" or to reading 808 from /proc/kmsg. 809 810 consoleblank= [KNL] The console blank (screen saver) timeout in 811 seconds. A value of 0 disables the blank timer. 812 Defaults to 0. 813 814 coredump_filter= 815 [KNL] Change the default value for 816 /proc/<pid>/coredump_filter. 817 See also Documentation/filesystems/proc.rst. 818 819 coresight_cpu_debug.enable 820 [ARM,ARM64] 821 Format: <bool> 822 Enable/disable the CPU sampling based debugging. 823 0: default value, disable debugging 824 1: enable debugging at boot time 825 826 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver 827 Format: 828 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>] 829 830 cpuidle.off=1 [CPU_IDLE] 831 disable the cpuidle sub-system 832 833 cpuidle.governor= 834 [CPU_IDLE] Name of the cpuidle governor to use. 835 836 cpufreq.off=1 [CPU_FREQ] 837 disable the cpufreq sub-system 838 839 cpufreq.default_governor= 840 [CPU_FREQ] Name of the default cpufreq governor or 841 policy to use. This governor must be registered in the 842 kernel before the cpufreq driver probes. 843 844 cpu_init_udelay=N 845 [X86] Delay for N microsec between assert and de-assert 846 of APIC INIT to start processors. This delay occurs 847 on every CPU online, such as boot, and resume from suspend. 848 Default: 10000 849 850 cpuhp.parallel= 851 [SMP] Enable/disable parallel bringup of secondary CPUs 852 Format: <bool> 853 Default is enabled if CONFIG_HOTPLUG_PARALLEL=y. Otherwise 854 the parameter has no effect. 855 856 crash_kexec_post_notifiers 857 Run kdump after running panic-notifiers and dumping 858 kmsg. This only for the users who doubt kdump always 859 succeeds in any situation. 860 Note that this also increases risks of kdump failure, 861 because some panic notifiers can make the crashed 862 kernel more unstable. 863 864 crashkernel=size[KMG][@offset[KMG]] 865 [KNL] Using kexec, Linux can switch to a 'crash kernel' 866 upon panic. This parameter reserves the physical 867 memory region [offset, offset + size] for that kernel 868 image. If '@offset' is omitted, then a suitable offset 869 is selected automatically. 870 [KNL, X86-64, ARM64, RISCV] Select a region under 4G first, and 871 fall back to reserve region above 4G when '@offset' 872 hasn't been specified. 873 See Documentation/admin-guide/kdump/kdump.rst for further details. 874 875 crashkernel=range1:size1[,range2:size2,...][@offset] 876 [KNL] Same as above, but depends on the memory 877 in the running system. The syntax of range is 878 start-[end] where start and end are both 879 a memory unit (amount[KMG]). See also 880 Documentation/admin-guide/kdump/kdump.rst for an example. 881 882 crashkernel=size[KMG],high 883 [KNL, X86-64, ARM64, RISCV] range could be above 4G. 884 Allow kernel to allocate physical memory region from top, 885 so could be above 4G if system have more than 4G ram 886 installed. Otherwise memory region will be allocated 887 below 4G, if available. 888 It will be ignored if crashkernel=X is specified. 889 crashkernel=size[KMG],low 890 [KNL, X86-64, ARM64, RISCV] range under 4G. When crashkernel=X,high 891 is passed, kernel could allocate physical memory region 892 above 4G, that cause second kernel crash on system 893 that require some amount of low memory, e.g. swiotlb 894 requires at least 64M+32K low memory, also enough extra 895 low memory is needed to make sure DMA buffers for 32-bit 896 devices won't run out. Kernel would try to allocate 897 default size of memory below 4G automatically. The default 898 size is platform dependent. 899 --> x86: max(swiotlb_size_or_default() + 8MiB, 256MiB) 900 --> arm64: 128MiB 901 --> riscv: 128MiB 902 This one lets the user specify own low range under 4G 903 for second kernel instead. 904 0: to disable low allocation. 905 It will be ignored when crashkernel=X,high is not used 906 or memory reserved is below 4G. 907 908 cryptomgr.notests 909 [KNL] Disable crypto self-tests 910 911 cs89x0_dma= [HW,NET] 912 Format: <dma> 913 914 cs89x0_media= [HW,NET] 915 Format: { rj45 | aui | bnc } 916 917 csdlock_debug= [KNL] Enable or disable debug add-ons of cross-CPU 918 function call handling. When switched on, 919 additional debug data is printed to the console 920 in case a hanging CPU is detected, and that 921 CPU is pinged again in order to try to resolve 922 the hang situation. The default value of this 923 option depends on the CSD_LOCK_WAIT_DEBUG_DEFAULT 924 Kconfig option. 925 926 dasd= [HW,NET] 927 See header of drivers/s390/block/dasd_devmap.c. 928 929 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port 930 (one device per port) 931 Format: <port#>,<type> 932 See also Documentation/input/devices/joystick-parport.rst 933 934 debug [KNL] Enable kernel debugging (events log level). 935 936 debug_boot_weak_hash 937 [KNL] Enable printing [hashed] pointers early in the 938 boot sequence. If enabled, we use a weak hash instead 939 of siphash to hash pointers. Use this option if you are 940 seeing instances of '(___ptrval___)') and need to see a 941 value (hashed pointer) instead. Cryptographically 942 insecure, please do not use on production kernels. 943 944 debug_locks_verbose= 945 [KNL] verbose locking self-tests 946 Format: <int> 947 Print debugging info while doing the locking API 948 self-tests. 949 Bitmask for the various LOCKTYPE_ tests. Defaults to 0 950 (no extra messages), setting it to -1 (all bits set) 951 will print _a_lot_ more information - normally only 952 useful to lockdep developers. 953 954 debug_objects [KNL] Enable object debugging 955 956 debug_guardpage_minorder= 957 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this 958 parameter allows control of the order of pages that will 959 be intentionally kept free (and hence protected) by the 960 buddy allocator. Bigger value increase the probability 961 of catching random memory corruption, but reduce the 962 amount of memory for normal system use. The maximum 963 possible value is MAX_ORDER/2. Setting this parameter 964 to 1 or 2 should be enough to identify most random 965 memory corruption problems caused by bugs in kernel or 966 driver code when a CPU writes to (or reads from) a 967 random memory location. Note that there exists a class 968 of memory corruptions problems caused by buggy H/W or 969 F/W or by drivers badly programming DMA (basically when 970 memory is written at bus level and the CPU MMU is 971 bypassed) which are not detectable by 972 CONFIG_DEBUG_PAGEALLOC, hence this option will not help 973 tracking down these problems. 974 975 debug_pagealloc= 976 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this parameter 977 enables the feature at boot time. By default, it is 978 disabled and the system will work mostly the same as a 979 kernel built without CONFIG_DEBUG_PAGEALLOC. 980 Note: to get most of debug_pagealloc error reports, it's 981 useful to also enable the page_owner functionality. 982 on: enable the feature 983 984 debugfs= [KNL] This parameter enables what is exposed to userspace 985 and debugfs internal clients. 986 Format: { on, no-mount, off } 987 on: All functions are enabled. 988 no-mount: 989 Filesystem is not registered but kernel clients can 990 access APIs and a crashkernel can be used to read 991 its content. There is nothing to mount. 992 off: Filesystem is not registered and clients 993 get a -EPERM as result when trying to register files 994 or directories within debugfs. 995 This is equivalent of the runtime functionality if 996 debugfs was not enabled in the kernel at all. 997 Default value is set in build-time with a kernel configuration. 998 999 debugpat [X86] Enable PAT debugging 1000 1001 default_hugepagesz= 1002 [HW] The size of the default HugeTLB page. This is 1003 the size represented by the legacy /proc/ hugepages 1004 APIs. In addition, this is the default hugetlb size 1005 used for shmget(), mmap() and mounting hugetlbfs 1006 filesystems. If not specified, defaults to the 1007 architecture's default huge page size. Huge page 1008 sizes are architecture dependent. See also 1009 Documentation/admin-guide/mm/hugetlbpage.rst. 1010 Format: size[KMG] 1011 1012 deferred_probe_timeout= 1013 [KNL] Debugging option to set a timeout in seconds for 1014 deferred probe to give up waiting on dependencies to 1015 probe. Only specific dependencies (subsystems or 1016 drivers) that have opted in will be ignored. A timeout 1017 of 0 will timeout at the end of initcalls. If the time 1018 out hasn't expired, it'll be restarted by each 1019 successful driver registration. This option will also 1020 dump out devices still on the deferred probe list after 1021 retrying. 1022 1023 delayacct [KNL] Enable per-task delay accounting 1024 1025 dell_smm_hwmon.ignore_dmi= 1026 [HW] Continue probing hardware even if DMI data 1027 indicates that the driver is running on unsupported 1028 hardware. 1029 1030 dell_smm_hwmon.force= 1031 [HW] Activate driver even if SMM BIOS signature does 1032 not match list of supported models and enable otherwise 1033 blacklisted features. 1034 1035 dell_smm_hwmon.power_status= 1036 [HW] Report power status in /proc/i8k 1037 (disabled by default). 1038 1039 dell_smm_hwmon.restricted= 1040 [HW] Allow controlling fans only if SYS_ADMIN 1041 capability is set. 1042 1043 dell_smm_hwmon.fan_mult= 1044 [HW] Factor to multiply fan speed with. 1045 1046 dell_smm_hwmon.fan_max= 1047 [HW] Maximum configurable fan speed. 1048 1049 dfltcc= [HW,S390] 1050 Format: { on | off | def_only | inf_only | always } 1051 on: s390 zlib hardware support for compression on 1052 level 1 and decompression (default) 1053 off: No s390 zlib hardware support 1054 def_only: s390 zlib hardware support for deflate 1055 only (compression on level 1) 1056 inf_only: s390 zlib hardware support for inflate 1057 only (decompression) 1058 always: Same as 'on' but ignores the selected compression 1059 level always using hardware support (used for debugging) 1060 1061 dhash_entries= [KNL] 1062 Set number of hash buckets for dentry cache. 1063 1064 disable_1tb_segments [PPC] 1065 Disables the use of 1TB hash page table segments. This 1066 causes the kernel to fall back to 256MB segments which 1067 can be useful when debugging issues that require an SLB 1068 miss to occur. 1069 1070 disable= [IPV6] 1071 See Documentation/networking/ipv6.rst. 1072 1073 disable_radix [PPC] 1074 Disable RADIX MMU mode on POWER9 1075 1076 disable_tlbie [PPC] 1077 Disable TLBIE instruction. Currently does not work 1078 with KVM, with HASH MMU, or with coherent accelerators. 1079 1080 disable_cpu_apicid= [X86,APIC,SMP] 1081 Format: <int> 1082 The number of initial APIC ID for the 1083 corresponding CPU to be disabled at boot, 1084 mostly used for the kdump 2nd kernel to 1085 disable BSP to wake up multiple CPUs without 1086 causing system reset or hang due to sending 1087 INIT from AP to BSP. 1088 1089 disable_ddw [PPC/PSERIES] 1090 Disable Dynamic DMA Window support. Use this 1091 to workaround buggy firmware. 1092 1093 disable_ipv6= [IPV6] 1094 See Documentation/networking/ipv6.rst. 1095 1096 disable_mtrr_cleanup [X86] 1097 The kernel tries to adjust MTRR layout from continuous 1098 to discrete, to make X server driver able to add WB 1099 entry later. This parameter disables that. 1100 1101 disable_mtrr_trim [X86, Intel and AMD only] 1102 By default the kernel will trim any uncacheable 1103 memory out of your available memory pool based on 1104 MTRR settings. This parameter disables that behavior, 1105 possibly causing your machine to run very slowly. 1106 1107 disable_timer_pin_1 [X86] 1108 Disable PIN 1 of APIC timer 1109 Can be useful to work around chipset bugs. 1110 1111 dis_ucode_ldr [X86] Disable the microcode loader. 1112 1113 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support, 1114 this option disables the debugging code at boot. 1115 1116 dma_debug_entries=<number> 1117 This option allows to tune the number of preallocated 1118 entries for DMA-API debugging code. One entry is 1119 required per DMA-API allocation. Use this if the 1120 DMA-API debugging code disables itself because the 1121 architectural default is too low. 1122 1123 dma_debug_driver=<driver_name> 1124 With this option the DMA-API debugging driver 1125 filter feature can be enabled at boot time. Just 1126 pass the driver to filter for as the parameter. 1127 The filter can be disabled or changed to another 1128 driver later using sysfs. 1129 1130 reg_file_data_sampling= 1131 [X86] Controls mitigation for Register File Data 1132 Sampling (RFDS) vulnerability. RFDS is a CPU 1133 vulnerability which may allow userspace to infer 1134 kernel data values previously stored in floating point 1135 registers, vector registers, or integer registers. 1136 RFDS only affects Intel Atom processors. 1137 1138 on: Turns ON the mitigation. 1139 off: Turns OFF the mitigation. 1140 1141 This parameter overrides the compile time default set 1142 by CONFIG_MITIGATION_RFDS. Mitigation cannot be 1143 disabled when other VERW based mitigations (like MDS) 1144 are enabled. In order to disable RFDS mitigation all 1145 VERW based mitigations need to be disabled. 1146 1147 For details see: 1148 Documentation/admin-guide/hw-vuln/reg-file-data-sampling.rst 1149 1150 driver_async_probe= [KNL] 1151 List of driver names to be probed asynchronously. * 1152 matches with all driver names. If * is specified, the 1153 rest of the listed driver names are those that will NOT 1154 match the *. 1155 Format: <driver_name1>,<driver_name2>... 1156 1157 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>] 1158 Broken monitors, graphic adapters, KVMs and EDIDless 1159 panels may send no or incorrect EDID data sets. 1160 This parameter allows to specify an EDID data sets 1161 in the /lib/firmware directory that are used instead. 1162 Generic built-in EDID data sets are used, if one of 1163 edid/1024x768.bin, edid/1280x1024.bin, 1164 edid/1680x1050.bin, or edid/1920x1080.bin is given 1165 and no file with the same name exists. Details and 1166 instructions how to build your own EDID data are 1167 available in Documentation/admin-guide/edid.rst. An EDID 1168 data set will only be used for a particular connector, 1169 if its name and a colon are prepended to the EDID 1170 name. Each connector may use a unique EDID data 1171 set by separating the files with a comma. An EDID 1172 data set with no connector name will be used for 1173 any connectors not explicitly specified. 1174 1175 dscc4.setup= [NET] 1176 1177 dt_cpu_ftrs= [PPC] 1178 Format: {"off" | "known"} 1179 Control how the dt_cpu_ftrs device-tree binding is 1180 used for CPU feature discovery and setup (if it 1181 exists). 1182 off: Do not use it, fall back to legacy cpu table. 1183 known: Do not pass through unknown features to guests 1184 or userspace, only those that the kernel is aware of. 1185 1186 dump_apple_properties [X86] 1187 Dump name and content of EFI device properties on 1188 x86 Macs. Useful for driver authors to determine 1189 what data is available or for reverse-engineering. 1190 1191 dyndbg[="val"] [KNL,DYNAMIC_DEBUG] 1192 <module>.dyndbg[="val"] 1193 Enable debug messages at boot time. See 1194 Documentation/admin-guide/dynamic-debug-howto.rst 1195 for details. 1196 1197 early_ioremap_debug [KNL] 1198 Enable debug messages in early_ioremap support. This 1199 is useful for tracking down temporary early mappings 1200 which are not unmapped. 1201 1202 earlycon= [KNL] Output early console device and options. 1203 1204 When used with no options, the early console is 1205 determined by stdout-path property in device tree's 1206 chosen node or the ACPI SPCR table if supported by 1207 the platform. 1208 1209 cdns,<addr>[,options] 1210 Start an early, polled-mode console on a Cadence 1211 (xuartps) serial port at the specified address. Only 1212 supported option is baud rate. If baud rate is not 1213 specified, the serial port must already be setup and 1214 configured. 1215 1216 uart[8250],io,<addr>[,options[,uartclk]] 1217 uart[8250],mmio,<addr>[,options[,uartclk]] 1218 uart[8250],mmio32,<addr>[,options[,uartclk]] 1219 uart[8250],mmio32be,<addr>[,options[,uartclk]] 1220 uart[8250],0x<addr>[,options] 1221 Start an early, polled-mode console on the 8250/16550 1222 UART at the specified I/O port or MMIO address. 1223 MMIO inter-register address stride is either 8-bit 1224 (mmio) or 32-bit (mmio32 or mmio32be). 1225 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed 1226 to be equivalent to 'mmio'. 'options' are specified 1227 in the same format described for "console=ttyS<n>"; if 1228 unspecified, the h/w is not initialized. 'uartclk' is 1229 the uart clock frequency; if unspecified, it is set 1230 to 'BASE_BAUD' * 16. 1231 1232 pl011,<addr> 1233 pl011,mmio32,<addr> 1234 Start an early, polled-mode console on a pl011 serial 1235 port at the specified address. The pl011 serial port 1236 must already be setup and configured. Options are not 1237 yet supported. If 'mmio32' is specified, then only 1238 the driver will use only 32-bit accessors to read/write 1239 the device registers. 1240 1241 liteuart,<addr> 1242 Start an early console on a litex serial port at the 1243 specified address. The serial port must already be 1244 setup and configured. Options are not yet supported. 1245 1246 meson,<addr> 1247 Start an early, polled-mode console on a meson serial 1248 port at the specified address. The serial port must 1249 already be setup and configured. Options are not yet 1250 supported. 1251 1252 msm_serial,<addr> 1253 Start an early, polled-mode console on an msm serial 1254 port at the specified address. The serial port 1255 must already be setup and configured. Options are not 1256 yet supported. 1257 1258 msm_serial_dm,<addr> 1259 Start an early, polled-mode console on an msm serial 1260 dm port at the specified address. The serial port 1261 must already be setup and configured. Options are not 1262 yet supported. 1263 1264 owl,<addr> 1265 Start an early, polled-mode console on a serial port 1266 of an Actions Semi SoC, such as S500 or S900, at the 1267 specified address. The serial port must already be 1268 setup and configured. Options are not yet supported. 1269 1270 rda,<addr> 1271 Start an early, polled-mode console on a serial port 1272 of an RDA Micro SoC, such as RDA8810PL, at the 1273 specified address. The serial port must already be 1274 setup and configured. Options are not yet supported. 1275 1276 sbi 1277 Use RISC-V SBI (Supervisor Binary Interface) for early 1278 console. 1279 1280 smh Use ARM semihosting calls for early console. 1281 1282 s3c2410,<addr> 1283 s3c2412,<addr> 1284 s3c2440,<addr> 1285 s3c6400,<addr> 1286 s5pv210,<addr> 1287 exynos4210,<addr> 1288 Use early console provided by serial driver available 1289 on Samsung SoCs, requires selecting proper type and 1290 a correct base address of the selected UART port. The 1291 serial port must already be setup and configured. 1292 Options are not yet supported. 1293 1294 lantiq,<addr> 1295 Start an early, polled-mode console on a lantiq serial 1296 (lqasc) port at the specified address. The serial port 1297 must already be setup and configured. Options are not 1298 yet supported. 1299 1300 lpuart,<addr> 1301 lpuart32,<addr> 1302 Use early console provided by Freescale LP UART driver 1303 found on Freescale Vybrid and QorIQ LS1021A processors. 1304 A valid base address must be provided, and the serial 1305 port must already be setup and configured. 1306 1307 ec_imx21,<addr> 1308 ec_imx6q,<addr> 1309 Start an early, polled-mode, output-only console on the 1310 Freescale i.MX UART at the specified address. The UART 1311 must already be setup and configured. 1312 1313 ar3700_uart,<addr> 1314 Start an early, polled-mode console on the 1315 Armada 3700 serial port at the specified 1316 address. The serial port must already be setup 1317 and configured. Options are not yet supported. 1318 1319 qcom_geni,<addr> 1320 Start an early, polled-mode console on a Qualcomm 1321 Generic Interface (GENI) based serial port at the 1322 specified address. The serial port must already be 1323 setup and configured. Options are not yet supported. 1324 1325 efifb,[options] 1326 Start an early, unaccelerated console on the EFI 1327 memory mapped framebuffer (if available). On cache 1328 coherent non-x86 systems that use system memory for 1329 the framebuffer, pass the 'ram' option so that it is 1330 mapped with the correct attributes. 1331 1332 linflex,<addr> 1333 Use early console provided by Freescale LINFlexD UART 1334 serial driver for NXP S32V234 SoCs. A valid base 1335 address must be provided, and the serial port must 1336 already be setup and configured. 1337 1338 earlyprintk= [X86,SH,ARM,M68k,S390] 1339 earlyprintk=vga 1340 earlyprintk=sclp 1341 earlyprintk=xen 1342 earlyprintk=serial[,ttySn[,baudrate]] 1343 earlyprintk=serial[,0x...[,baudrate]] 1344 earlyprintk=ttySn[,baudrate] 1345 earlyprintk=dbgp[debugController#] 1346 earlyprintk=pciserial[,force],bus:device.function[,baudrate] 1347 earlyprintk=xdbc[xhciController#] 1348 1349 earlyprintk is useful when the kernel crashes before 1350 the normal console is initialized. It is not enabled by 1351 default because it has some cosmetic problems. 1352 1353 Append ",keep" to not disable it when the real console 1354 takes over. 1355 1356 Only one of vga, serial, or usb debug port can 1357 be used at a time. 1358 1359 Currently only ttyS0 and ttyS1 may be specified by 1360 name. Other I/O ports may be explicitly specified 1361 on some architectures (x86 and arm at least) by 1362 replacing ttySn with an I/O port address, like this: 1363 earlyprintk=serial,0x1008,115200 1364 You can find the port for a given device in 1365 /proc/tty/driver/serial: 1366 2: uart:ST16650V2 port:00001008 irq:18 ... 1367 1368 Interaction with the standard serial driver is not 1369 very good. 1370 1371 The VGA output is eventually overwritten by 1372 the real console. 1373 1374 The xen option can only be used in Xen domains. 1375 1376 The sclp output can only be used on s390. 1377 1378 The optional "force" to "pciserial" enables use of a 1379 PCI device even when its classcode is not of the 1380 UART class. 1381 1382 edac_report= [HW,EDAC] Control how to report EDAC event 1383 Format: {"on" | "off" | "force"} 1384 on: enable EDAC to report H/W event. May be overridden 1385 by other higher priority error reporting module. 1386 off: disable H/W event reporting through EDAC. 1387 force: enforce the use of EDAC to report H/W event. 1388 default: on. 1389 1390 edd= [EDD] 1391 Format: {"off" | "on" | "skip[mbr]"} 1392 1393 efi= [EFI] 1394 Format: { "debug", "disable_early_pci_dma", 1395 "nochunk", "noruntime", "nosoftreserve", 1396 "novamap", "no_disable_early_pci_dma" } 1397 debug: enable misc debug output. 1398 disable_early_pci_dma: disable the busmaster bit on all 1399 PCI bridges while in the EFI boot stub. 1400 nochunk: disable reading files in "chunks" in the EFI 1401 boot stub, as chunking can cause problems with some 1402 firmware implementations. 1403 noruntime : disable EFI runtime services support 1404 nosoftreserve: The EFI_MEMORY_SP (Specific Purpose) 1405 attribute may cause the kernel to reserve the 1406 memory range for a memory mapping driver to 1407 claim. Specify efi=nosoftreserve to disable this 1408 reservation and treat the memory by its base type 1409 (i.e. EFI_CONVENTIONAL_MEMORY / "System RAM"). 1410 novamap: do not call SetVirtualAddressMap(). 1411 no_disable_early_pci_dma: Leave the busmaster bit set 1412 on all PCI bridges while in the EFI boot stub 1413 1414 efi_no_storage_paranoia [EFI; X86] 1415 Using this parameter you can use more than 50% of 1416 your efi variable storage. Use this parameter only if 1417 you are really sure that your UEFI does sane gc and 1418 fulfills the spec otherwise your board may brick. 1419 1420 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86] 1421 Add arbitrary attribute to specific memory range by 1422 updating original EFI memory map. 1423 Region of memory which aa attribute is added to is 1424 from ss to ss+nn. 1425 1426 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000 1427 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000) 1428 attribute is added to range 0x100000000-0x180000000 and 1429 0x10a0000000-0x1120000000. 1430 1431 If efi_fake_mem=8G@9G:0x40000 is specified, the 1432 EFI_MEMORY_SP(0x40000) attribute is added to 1433 range 0x240000000-0x43fffffff. 1434 1435 Using this parameter you can do debugging of EFI memmap 1436 related features. For example, you can do debugging of 1437 Address Range Mirroring feature even if your box 1438 doesn't support it, or mark specific memory as 1439 "soft reserved". 1440 1441 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT 1442 that is to be dynamically loaded by Linux. If there are 1443 multiple variables with the same name but with different 1444 vendor GUIDs, all of them will be loaded. See 1445 Documentation/admin-guide/acpi/ssdt-overlays.rst for details. 1446 1447 1448 eisa_irq_edge= [PARISC,HW] 1449 See header of drivers/parisc/eisa.c. 1450 1451 ekgdboc= [X86,KGDB] Allow early kernel console debugging 1452 Format: ekgdboc=kbd 1453 1454 This is designed to be used in conjunction with 1455 the boot argument: earlyprintk=vga 1456 1457 This parameter works in place of the kgdboc parameter 1458 but can only be used if the backing tty is available 1459 very early in the boot process. For early debugging 1460 via a serial port see kgdboc_earlycon instead. 1461 1462 elanfreq= [X86-32] 1463 See comment before function elanfreq_setup() in 1464 arch/x86/kernel/cpu/cpufreq/elanfreq.c. 1465 1466 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390] 1467 Specifies physical address of start of kernel core 1468 image elf header and optionally the size. Generally 1469 kexec loader will pass this option to capture kernel. 1470 See Documentation/admin-guide/kdump/kdump.rst for details. 1471 1472 enable_mtrr_cleanup [X86] 1473 The kernel tries to adjust MTRR layout from continuous 1474 to discrete, to make X server driver able to add WB 1475 entry later. This parameter enables that. 1476 1477 enable_timer_pin_1 [X86] 1478 Enable PIN 1 of APIC timer 1479 Can be useful to work around chipset bugs 1480 (in particular on some ATI chipsets). 1481 The kernel tries to set a reasonable default. 1482 1483 enforcing= [SELINUX] Set initial enforcing status. 1484 Format: {"0" | "1"} 1485 See security/selinux/Kconfig help text. 1486 0 -- permissive (log only, no denials). 1487 1 -- enforcing (deny and log). 1488 Default value is 0. 1489 Value can be changed at runtime via 1490 /sys/fs/selinux/enforce. 1491 1492 erst_disable [ACPI] 1493 Disable Error Record Serialization Table (ERST) 1494 support. 1495 1496 ether= [HW,NET] Ethernet cards parameters 1497 This option is obsoleted by the "netdev=" option, which 1498 has equivalent usage. See its documentation for details. 1499 1500 evm= [EVM] 1501 Format: { "fix" } 1502 Permit 'security.evm' to be updated regardless of 1503 current integrity status. 1504 1505 early_page_ext [KNL] Enforces page_ext initialization to earlier 1506 stages so cover more early boot allocations. 1507 Please note that as side effect some optimizations 1508 might be disabled to achieve that (e.g. parallelized 1509 memory initialization is disabled) so the boot process 1510 might take longer, especially on systems with a lot of 1511 memory. Available with CONFIG_PAGE_EXTENSION=y. 1512 1513 failslab= 1514 fail_usercopy= 1515 fail_page_alloc= 1516 fail_make_request=[KNL] 1517 General fault injection mechanism. 1518 Format: <interval>,<probability>,<space>,<times> 1519 See also Documentation/fault-injection/. 1520 1521 fb_tunnels= [NET] 1522 Format: { initns | none } 1523 See Documentation/admin-guide/sysctl/net.rst for 1524 fb_tunnels_only_for_init_ns 1525 1526 floppy= [HW] 1527 See Documentation/admin-guide/blockdev/floppy.rst. 1528 1529 force_pal_cache_flush 1530 [IA-64] Avoid check_sal_cache_flush which may hang on 1531 buggy SAL_CACHE_FLUSH implementations. Using this 1532 parameter will force ia64_sal_cache_flush to call 1533 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH. 1534 1535 forcepae [X86-32] 1536 Forcefully enable Physical Address Extension (PAE). 1537 Many Pentium M systems disable PAE but may have a 1538 functionally usable PAE implementation. 1539 Warning: use of this parameter will taint the kernel 1540 and may cause unknown problems. 1541 1542 ftrace=[tracer] 1543 [FTRACE] will set and start the specified tracer 1544 as early as possible in order to facilitate early 1545 boot debugging. 1546 1547 ftrace_boot_snapshot 1548 [FTRACE] On boot up, a snapshot will be taken of the 1549 ftrace ring buffer that can be read at: 1550 /sys/kernel/tracing/snapshot. 1551 This is useful if you need tracing information from kernel 1552 boot up that is likely to be overridden by user space 1553 start up functionality. 1554 1555 Optionally, the snapshot can also be defined for a tracing 1556 instance that was created by the trace_instance= command 1557 line parameter. 1558 1559 trace_instance=foo,sched_switch ftrace_boot_snapshot=foo 1560 1561 The above will cause the "foo" tracing instance to trigger 1562 a snapshot at the end of boot up. 1563 1564 ftrace_dump_on_oops[=orig_cpu] 1565 [FTRACE] will dump the trace buffers on oops. 1566 If no parameter is passed, ftrace will dump 1567 buffers of all CPUs, but if you pass orig_cpu, it will 1568 dump only the buffer of the CPU that triggered the 1569 oops. 1570 1571 ftrace_filter=[function-list] 1572 [FTRACE] Limit the functions traced by the function 1573 tracer at boot up. function-list is a comma-separated 1574 list of functions. This list can be changed at run 1575 time by the set_ftrace_filter file in the debugfs 1576 tracing directory. 1577 1578 ftrace_notrace=[function-list] 1579 [FTRACE] Do not trace the functions specified in 1580 function-list. This list can be changed at run time 1581 by the set_ftrace_notrace file in the debugfs 1582 tracing directory. 1583 1584 ftrace_graph_filter=[function-list] 1585 [FTRACE] Limit the top level callers functions traced 1586 by the function graph tracer at boot up. 1587 function-list is a comma-separated list of functions 1588 that can be changed at run time by the 1589 set_graph_function file in the debugfs tracing directory. 1590 1591 ftrace_graph_notrace=[function-list] 1592 [FTRACE] Do not trace from the functions specified in 1593 function-list. This list is a comma-separated list of 1594 functions that can be changed at run time by the 1595 set_graph_notrace file in the debugfs tracing directory. 1596 1597 ftrace_graph_max_depth=<uint> 1598 [FTRACE] Used with the function graph tracer. This is 1599 the max depth it will trace into a function. This value 1600 can be changed at run time by the max_graph_depth file 1601 in the tracefs tracing directory. default: 0 (no limit) 1602 1603 fw_devlink= [KNL] Create device links between consumer and supplier 1604 devices by scanning the firmware to infer the 1605 consumer/supplier relationships. This feature is 1606 especially useful when drivers are loaded as modules as 1607 it ensures proper ordering of tasks like device probing 1608 (suppliers first, then consumers), supplier boot state 1609 clean up (only after all consumers have probed), 1610 suspend/resume & runtime PM (consumers first, then 1611 suppliers). 1612 Format: { off | permissive | on | rpm } 1613 off -- Don't create device links from firmware info. 1614 permissive -- Create device links from firmware info 1615 but use it only for ordering boot state clean 1616 up (sync_state() calls). 1617 on -- Create device links from firmware info and use it 1618 to enforce probe and suspend/resume ordering. 1619 rpm -- Like "on", but also use to order runtime PM. 1620 1621 fw_devlink.strict=<bool> 1622 [KNL] Treat all inferred dependencies as mandatory 1623 dependencies. This only applies for fw_devlink=on|rpm. 1624 Format: <bool> 1625 1626 fw_devlink.sync_state = 1627 [KNL] When all devices that could probe have finished 1628 probing, this parameter controls what to do with 1629 devices that haven't yet received their sync_state() 1630 calls. 1631 Format: { strict | timeout } 1632 strict -- Default. Continue waiting on consumers to 1633 probe successfully. 1634 timeout -- Give up waiting on consumers and call 1635 sync_state() on any devices that haven't yet 1636 received their sync_state() calls after 1637 deferred_probe_timeout has expired or by 1638 late_initcall() if !CONFIG_MODULES. 1639 1640 gamecon.map[2|3]= 1641 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad 1642 support via parallel port (up to 5 devices per port) 1643 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5> 1644 See also Documentation/input/devices/joystick-parport.rst 1645 1646 gamma= [HW,DRM] 1647 1648 gart_fix_e820= [X86-64] disable the fix e820 for K8 GART 1649 Format: off | on 1650 default: on 1651 1652 gather_data_sampling= 1653 [X86,INTEL] Control the Gather Data Sampling (GDS) 1654 mitigation. 1655 1656 Gather Data Sampling is a hardware vulnerability which 1657 allows unprivileged speculative access to data which was 1658 previously stored in vector registers. 1659 1660 This issue is mitigated by default in updated microcode. 1661 The mitigation may have a performance impact but can be 1662 disabled. On systems without the microcode mitigation 1663 disabling AVX serves as a mitigation. 1664 1665 force: Disable AVX to mitigate systems without 1666 microcode mitigation. No effect if the microcode 1667 mitigation is present. Known to cause crashes in 1668 userspace with buggy AVX enumeration. 1669 1670 off: Disable GDS mitigation. 1671 1672 gcov_persist= [GCOV] When non-zero (default), profiling data for 1673 kernel modules is saved and remains accessible via 1674 debugfs, even when the module is unloaded/reloaded. 1675 When zero, profiling data is discarded and associated 1676 debugfs files are removed at module unload time. 1677 1678 goldfish [X86] Enable the goldfish android emulator platform. 1679 Don't use this when you are not running on the 1680 android emulator 1681 1682 gpio-mockup.gpio_mockup_ranges 1683 [HW] Sets the ranges of gpiochip of for this device. 1684 Format: <start1>,<end1>,<start2>,<end2>... 1685 gpio-mockup.gpio_mockup_named_lines 1686 [HW] Let the driver know GPIO lines should be named. 1687 1688 gpt [EFI] Forces disk with valid GPT signature but 1689 invalid Protective MBR to be treated as GPT. If the 1690 primary GPT is corrupted, it enables the backup/alternate 1691 GPT to be used instead. 1692 1693 grcan.enable0= [HW] Configuration of physical interface 0. Determines 1694 the "Enable 0" bit of the configuration register. 1695 Format: 0 | 1 1696 Default: 0 1697 grcan.enable1= [HW] Configuration of physical interface 1. Determines 1698 the "Enable 0" bit of the configuration register. 1699 Format: 0 | 1 1700 Default: 0 1701 grcan.select= [HW] Select which physical interface to use. 1702 Format: 0 | 1 1703 Default: 0 1704 grcan.txsize= [HW] Sets the size of the tx buffer. 1705 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0. 1706 Default: 1024 1707 grcan.rxsize= [HW] Sets the size of the rx buffer. 1708 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0. 1709 Default: 1024 1710 1711 hardened_usercopy= 1712 [KNL] Under CONFIG_HARDENED_USERCOPY, whether 1713 hardening is enabled for this boot. Hardened 1714 usercopy checking is used to protect the kernel 1715 from reading or writing beyond known memory 1716 allocation boundaries as a proactive defense 1717 against bounds-checking flaws in the kernel's 1718 copy_to_user()/copy_from_user() interface. 1719 on Perform hardened usercopy checks (default). 1720 off Disable hardened usercopy checks. 1721 1722 hardlockup_all_cpu_backtrace= 1723 [KNL] Should the hard-lockup detector generate 1724 backtraces on all cpus. 1725 Format: 0 | 1 1726 1727 hashdist= [KNL,NUMA] Large hashes allocated during boot 1728 are distributed across NUMA nodes. Defaults on 1729 for 64-bit NUMA, off otherwise. 1730 Format: 0 | 1 (for off | on) 1731 1732 hcl= [IA-64] SGI's Hardware Graph compatibility layer 1733 1734 hd= [EIDE] (E)IDE hard drive subsystem geometry 1735 Format: <cyl>,<head>,<sect> 1736 1737 hest_disable [ACPI] 1738 Disable Hardware Error Source Table (HEST) support; 1739 corresponding firmware-first mode error processing 1740 logic will be disabled. 1741 1742 hibernate= [HIBERNATION] 1743 noresume Don't check if there's a hibernation image 1744 present during boot. 1745 nocompress Don't compress/decompress hibernation images. 1746 no Disable hibernation and resume. 1747 protect_image Turn on image protection during restoration 1748 (that will set all pages holding image data 1749 during restoration read-only). 1750 1751 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact 1752 size of <nn>. This works even on boxes that have no 1753 highmem otherwise. This also works to reduce highmem 1754 size on bigger boxes. 1755 1756 highres= [KNL] Enable/disable high resolution timer mode. 1757 Valid parameters: "on", "off" 1758 Default: "on" 1759 1760 hlt [BUGS=ARM,SH] 1761 1762 hostname= [KNL] Set the hostname (aka UTS nodename). 1763 Format: <string> 1764 This allows setting the system's hostname during early 1765 startup. This sets the name returned by gethostname. 1766 Using this parameter to set the hostname makes it 1767 possible to ensure the hostname is correctly set before 1768 any userspace processes run, avoiding the possibility 1769 that a process may call gethostname before the hostname 1770 has been explicitly set, resulting in the calling 1771 process getting an incorrect result. The string must 1772 not exceed the maximum allowed hostname length (usually 1773 64 characters) and will be truncated otherwise. 1774 1775 hpet= [X86-32,HPET] option to control HPET usage 1776 Format: { enable (default) | disable | force | 1777 verbose } 1778 disable: disable HPET and use PIT instead 1779 force: allow force enabled of undocumented chips (ICH4, 1780 VIA, nVidia) 1781 verbose: show contents of HPET registers during setup 1782 1783 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET 1784 registers. Default set by CONFIG_HPET_MMAP_DEFAULT. 1785 1786 hugepages= [HW] Number of HugeTLB pages to allocate at boot. 1787 If this follows hugepagesz (below), it specifies 1788 the number of pages of hugepagesz to be allocated. 1789 If this is the first HugeTLB parameter on the command 1790 line, it specifies the number of pages to allocate for 1791 the default huge page size. If using node format, the 1792 number of pages to allocate per-node can be specified. 1793 See also Documentation/admin-guide/mm/hugetlbpage.rst. 1794 Format: <integer> or (node format) 1795 <node>:<integer>[,<node>:<integer>] 1796 1797 hugepagesz= 1798 [HW] The size of the HugeTLB pages. This is used in 1799 conjunction with hugepages (above) to allocate huge 1800 pages of a specific size at boot. The pair 1801 hugepagesz=X hugepages=Y can be specified once for 1802 each supported huge page size. Huge page sizes are 1803 architecture dependent. See also 1804 Documentation/admin-guide/mm/hugetlbpage.rst. 1805 Format: size[KMG] 1806 1807 hugetlb_cma= [HW,CMA] The size of a CMA area used for allocation 1808 of gigantic hugepages. Or using node format, the size 1809 of a CMA area per node can be specified. 1810 Format: nn[KMGTPE] or (node format) 1811 <node>:nn[KMGTPE][,<node>:nn[KMGTPE]] 1812 1813 Reserve a CMA area of given size and allocate gigantic 1814 hugepages using the CMA allocator. If enabled, the 1815 boot-time allocation of gigantic hugepages is skipped. 1816 1817 hugetlb_free_vmemmap= 1818 [KNL] Requires CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP 1819 enabled. 1820 Control if HugeTLB Vmemmap Optimization (HVO) is enabled. 1821 Allows heavy hugetlb users to free up some more 1822 memory (7 * PAGE_SIZE for each 2MB hugetlb page). 1823 Format: { on | off (default) } 1824 1825 on: enable HVO 1826 off: disable HVO 1827 1828 Built with CONFIG_HUGETLB_PAGE_OPTIMIZE_VMEMMAP_DEFAULT_ON=y, 1829 the default is on. 1830 1831 Note that the vmemmap pages may be allocated from the added 1832 memory block itself when memory_hotplug.memmap_on_memory is 1833 enabled, those vmemmap pages cannot be optimized even if this 1834 feature is enabled. Other vmemmap pages not allocated from 1835 the added memory block itself do not be affected. 1836 1837 hung_task_panic= 1838 [KNL] Should the hung task detector generate panics. 1839 Format: 0 | 1 1840 1841 A value of 1 instructs the kernel to panic when a 1842 hung task is detected. The default value is controlled 1843 by the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time 1844 option. The value selected by this boot parameter can 1845 be changed later by the kernel.hung_task_panic sysctl. 1846 1847 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC) 1848 terminal devices. Valid values: 0..8 1849 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs. 1850 If specified, z/VM IUCV HVC accepts connections 1851 from listed z/VM user IDs only. 1852 1853 hv_nopvspin [X86,HYPER_V] Disables the paravirt spinlock optimizations 1854 which allow the hypervisor to 'idle' the 1855 guest on lock contention. 1856 1857 i2c_bus= [HW] Override the default board specific I2C bus speed 1858 or register an additional I2C bus that is not 1859 registered from board initialization code. 1860 Format: 1861 <bus_id>,<clkrate> 1862 1863 i8042.debug [HW] Toggle i8042 debug mode 1864 i8042.unmask_kbd_data 1865 [HW] Enable printing of interrupt data from the KBD port 1866 (disabled by default, and as a pre-condition 1867 requires that i8042.debug=1 be enabled) 1868 i8042.direct [HW] Put keyboard port into non-translated mode 1869 i8042.dumbkbd [HW] Pretend that controller can only read data from 1870 keyboard and cannot control its state 1871 (Don't attempt to blink the leds) 1872 i8042.noaux [HW] Don't check for auxiliary (== mouse) port 1873 i8042.nokbd [HW] Don't check/create keyboard port 1874 i8042.noloop [HW] Disable the AUX Loopback command while probing 1875 for the AUX port 1876 i8042.nomux [HW] Don't check presence of an active multiplexing 1877 controller 1878 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX 1879 controllers 1880 i8042.notimeout [HW] Ignore timeout condition signalled by controller 1881 i8042.reset [HW] Reset the controller during init, cleanup and 1882 suspend-to-ram transitions, only during s2r 1883 transitions, or never reset 1884 Format: { 1 | Y | y | 0 | N | n } 1885 1, Y, y: always reset controller 1886 0, N, n: don't ever reset controller 1887 Default: only on s2r transitions on x86; most other 1888 architectures force reset to be always executed 1889 i8042.unlock [HW] Unlock (ignore) the keylock 1890 i8042.kbdreset [HW] Reset device connected to KBD port 1891 i8042.probe_defer 1892 [HW] Allow deferred probing upon i8042 probe errors 1893 1894 i810= [HW,DRM] 1895 1896 i915.invert_brightness= 1897 [DRM] Invert the sense of the variable that is used to 1898 set the brightness of the panel backlight. Normally a 1899 brightness value of 0 indicates backlight switched off, 1900 and the maximum of the brightness value sets the backlight 1901 to maximum brightness. If this parameter is set to 0 1902 (default) and the machine requires it, or this parameter 1903 is set to 1, a brightness value of 0 sets the backlight 1904 to maximum brightness, and the maximum of the brightness 1905 value switches the backlight off. 1906 -1 -- never invert brightness 1907 0 -- machine default 1908 1 -- force brightness inversion 1909 1910 icn= [HW,ISDN] 1911 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]] 1912 1913 1914 idle= [X86] 1915 Format: idle=poll, idle=halt, idle=nomwait 1916 Poll forces a polling idle loop that can slightly 1917 improve the performance of waking up a idle CPU, but 1918 will use a lot of power and make the system run hot. 1919 Not recommended. 1920 idle=halt: Halt is forced to be used for CPU idle. 1921 In such case C2/C3 won't be used again. 1922 idle=nomwait: Disable mwait for CPU C-states 1923 1924 idxd.sva= [HW] 1925 Format: <bool> 1926 Allow force disabling of Shared Virtual Memory (SVA) 1927 support for the idxd driver. By default it is set to 1928 true (1). 1929 1930 idxd.tc_override= [HW] 1931 Format: <bool> 1932 Allow override of default traffic class configuration 1933 for the device. By default it is set to false (0). 1934 1935 ieee754= [MIPS] Select IEEE Std 754 conformance mode 1936 Format: { strict | legacy | 2008 | relaxed } 1937 Default: strict 1938 1939 Choose which programs will be accepted for execution 1940 based on the IEEE 754 NaN encoding(s) supported by 1941 the FPU and the NaN encoding requested with the value 1942 of an ELF file header flag individually set by each 1943 binary. Hardware implementations are permitted to 1944 support either or both of the legacy and the 2008 NaN 1945 encoding mode. 1946 1947 Available settings are as follows: 1948 strict accept binaries that request a NaN encoding 1949 supported by the FPU 1950 legacy only accept legacy-NaN binaries, if supported 1951 by the FPU 1952 2008 only accept 2008-NaN binaries, if supported 1953 by the FPU 1954 relaxed accept any binaries regardless of whether 1955 supported by the FPU 1956 1957 The FPU emulator is always able to support both NaN 1958 encodings, so if no FPU hardware is present or it has 1959 been disabled with 'nofpu', then the settings of 1960 'legacy' and '2008' strap the emulator accordingly, 1961 'relaxed' straps the emulator for both legacy-NaN and 1962 2008-NaN, whereas 'strict' enables legacy-NaN only on 1963 legacy processors and both NaN encodings on MIPS32 or 1964 MIPS64 CPUs. 1965 1966 The setting for ABS.fmt/NEG.fmt instruction execution 1967 mode generally follows that for the NaN encoding, 1968 except where unsupported by hardware. 1969 1970 ignore_loglevel [KNL] 1971 Ignore loglevel setting - this will print /all/ 1972 kernel messages to the console. Useful for debugging. 1973 We also add it as printk module parameter, so users 1974 could change it dynamically, usually by 1975 /sys/module/printk/parameters/ignore_loglevel. 1976 1977 ignore_rlimit_data 1978 Ignore RLIMIT_DATA setting for data mappings, 1979 print warning at first misuse. Can be changed via 1980 /sys/module/kernel/parameters/ignore_rlimit_data. 1981 1982 ihash_entries= [KNL] 1983 Set number of hash buckets for inode cache. 1984 1985 ima_appraise= [IMA] appraise integrity measurements 1986 Format: { "off" | "enforce" | "fix" | "log" } 1987 default: "enforce" 1988 1989 ima_appraise_tcb [IMA] Deprecated. Use ima_policy= instead. 1990 The builtin appraise policy appraises all files 1991 owned by uid=0. 1992 1993 ima_canonical_fmt [IMA] 1994 Use the canonical format for the binary runtime 1995 measurements, instead of host native format. 1996 1997 ima_hash= [IMA] 1998 Format: { md5 | sha1 | rmd160 | sha256 | sha384 1999 | sha512 | ... } 2000 default: "sha1" 2001 2002 The list of supported hash algorithms is defined 2003 in crypto/hash_info.h. 2004 2005 ima_policy= [IMA] 2006 The builtin policies to load during IMA setup. 2007 Format: "tcb | appraise_tcb | secure_boot | 2008 fail_securely | critical_data" 2009 2010 The "tcb" policy measures all programs exec'd, files 2011 mmap'd for exec, and all files opened with the read 2012 mode bit set by either the effective uid (euid=0) or 2013 uid=0. 2014 2015 The "appraise_tcb" policy appraises the integrity of 2016 all files owned by root. 2017 2018 The "secure_boot" policy appraises the integrity 2019 of files (eg. kexec kernel image, kernel modules, 2020 firmware, policy, etc) based on file signatures. 2021 2022 The "fail_securely" policy forces file signature 2023 verification failure also on privileged mounted 2024 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE 2025 flag. 2026 2027 The "critical_data" policy measures kernel integrity 2028 critical data. 2029 2030 ima_tcb [IMA] Deprecated. Use ima_policy= instead. 2031 Load a policy which meets the needs of the Trusted 2032 Computing Base. This means IMA will measure all 2033 programs exec'd, files mmap'd for exec, and all files 2034 opened for read by uid=0. 2035 2036 ima_template= [IMA] 2037 Select one of defined IMA measurements template formats. 2038 Formats: { "ima" | "ima-ng" | "ima-ngv2" | "ima-sig" | 2039 "ima-sigv2" } 2040 Default: "ima-ng" 2041 2042 ima_template_fmt= 2043 [IMA] Define a custom template format. 2044 Format: { "field1|...|fieldN" } 2045 2046 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage 2047 Format: <min_file_size> 2048 Set the minimal file size for using asynchronous hash. 2049 If left unspecified, ahash usage is disabled. 2050 2051 ahash performance varies for different data sizes on 2052 different crypto accelerators. This option can be used 2053 to achieve the best performance for a particular HW. 2054 2055 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size 2056 Format: <bufsize> 2057 Set hashing buffer size. Default: 4k. 2058 2059 ahash performance varies for different chunk sizes on 2060 different crypto accelerators. This option can be used 2061 to achieve best performance for particular HW. 2062 2063 indirect_target_selection= [X86,Intel] Mitigation control for Indirect 2064 Target Selection(ITS) bug in Intel CPUs. Updated 2065 microcode is also required for a fix in IBPB. 2066 2067 on: Enable mitigation (default). 2068 off: Disable mitigation. 2069 force: Force the ITS bug and deploy default 2070 mitigation. 2071 vmexit: Only deploy mitigation if CPU is affected by 2072 guest/host isolation part of ITS. 2073 stuff: Deploy RSB-fill mitigation when retpoline is 2074 also deployed. Otherwise, deploy the default 2075 mitigation. 2076 2077 For details see: 2078 Documentation/admin-guide/hw-vuln/indirect-target-selection.rst 2079 2080 init= [KNL] 2081 Format: <full_path> 2082 Run specified binary instead of /sbin/init as init 2083 process. 2084 2085 initcall_debug [KNL] Trace initcalls as they are executed. Useful 2086 for working out where the kernel is dying during 2087 startup. 2088 2089 initcall_blacklist= [KNL] Do not execute a comma-separated list of 2090 initcall functions. Useful for debugging built-in 2091 modules and initcalls. 2092 2093 initramfs_async= [KNL] 2094 Format: <bool> 2095 Default: 1 2096 This parameter controls whether the initramfs 2097 image is unpacked asynchronously, concurrently 2098 with devices being probed and 2099 initialized. This should normally just work, 2100 but as a debugging aid, one can get the 2101 historical behaviour of the initramfs 2102 unpacking being completed before device_ and 2103 late_ initcalls. 2104 2105 initrd= [BOOT] Specify the location of the initial ramdisk 2106 2107 initrdmem= [KNL] Specify a physical address and size from which to 2108 load the initrd. If an initrd is compiled in or 2109 specified in the bootparams, it takes priority over this 2110 setting. 2111 Format: ss[KMG],nn[KMG] 2112 Default is 0, 0 2113 2114 init_on_alloc= [MM] Fill newly allocated pages and heap objects with 2115 zeroes. 2116 Format: 0 | 1 2117 Default set by CONFIG_INIT_ON_ALLOC_DEFAULT_ON. 2118 2119 init_on_free= [MM] Fill freed pages and heap objects with zeroes. 2120 Format: 0 | 1 2121 Default set by CONFIG_INIT_ON_FREE_DEFAULT_ON. 2122 2123 init_pkru= [X86] Specify the default memory protection keys rights 2124 register contents for all processes. 0x55555554 by 2125 default (disallow access to all but pkey 0). Can 2126 override in debugfs after boot. 2127 2128 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver 2129 Format: <irq> 2130 2131 int_pln_enable [X86] Enable power limit notification interrupt 2132 2133 integrity_audit=[IMA] 2134 Format: { "0" | "1" } 2135 0 -- basic integrity auditing messages. (Default) 2136 1 -- additional integrity auditing messages. 2137 2138 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option 2139 on 2140 Enable intel iommu driver. 2141 off 2142 Disable intel iommu driver. 2143 igfx_off [Default Off] 2144 By default, gfx is mapped as normal device. If a gfx 2145 device has a dedicated DMAR unit, the DMAR unit is 2146 bypassed by not enabling DMAR with this option. In 2147 this case, gfx device will use physical address for 2148 DMA. 2149 strict [Default Off] 2150 Deprecated, equivalent to iommu.strict=1. 2151 sp_off [Default Off] 2152 By default, super page will be supported if Intel IOMMU 2153 has the capability. With this option, super page will 2154 not be supported. 2155 sm_on 2156 Enable the Intel IOMMU scalable mode if the hardware 2157 advertises that it has support for the scalable mode 2158 translation. 2159 sm_off 2160 Disallow use of the Intel IOMMU scalable mode. 2161 tboot_noforce [Default Off] 2162 Do not force the Intel IOMMU enabled under tboot. 2163 By default, tboot will force Intel IOMMU on, which 2164 could harm performance of some high-throughput 2165 devices like 40GBit network cards, even if identity 2166 mapping is enabled. 2167 Note that using this option lowers the security 2168 provided by tboot because it makes the system 2169 vulnerable to DMA attacks. 2170 2171 intel_idle.max_cstate= [KNL,HW,ACPI,X86] 2172 0 disables intel_idle and fall back on acpi_idle. 2173 1 to 9 specify maximum depth of C-state. 2174 2175 intel_pstate= [X86] 2176 disable 2177 Do not enable intel_pstate as the default 2178 scaling driver for the supported processors 2179 active 2180 Use intel_pstate driver to bypass the scaling 2181 governors layer of cpufreq and provides it own 2182 algorithms for p-state selection. There are two 2183 P-state selection algorithms provided by 2184 intel_pstate in the active mode: powersave and 2185 performance. The way they both operate depends 2186 on whether or not the hardware managed P-states 2187 (HWP) feature has been enabled in the processor 2188 and possibly on the processor model. 2189 passive 2190 Use intel_pstate as a scaling driver, but configure it 2191 to work with generic cpufreq governors (instead of 2192 enabling its internal governor). This mode cannot be 2193 used along with the hardware-managed P-states (HWP) 2194 feature. 2195 force 2196 Enable intel_pstate on systems that prohibit it by default 2197 in favor of acpi-cpufreq. Forcing the intel_pstate driver 2198 instead of acpi-cpufreq may disable platform features, such 2199 as thermal controls and power capping, that rely on ACPI 2200 P-States information being indicated to OSPM and therefore 2201 should be used with caution. This option does not work with 2202 processors that aren't supported by the intel_pstate driver 2203 or on platforms that use pcc-cpufreq instead of acpi-cpufreq. 2204 no_hwp 2205 Do not enable hardware P state control (HWP) 2206 if available. 2207 hwp_only 2208 Only load intel_pstate on systems which support 2209 hardware P state control (HWP) if available. 2210 support_acpi_ppc 2211 Enforce ACPI _PPC performance limits. If the Fixed ACPI 2212 Description Table, specifies preferred power management 2213 profile as "Enterprise Server" or "Performance Server", 2214 then this feature is turned on by default. 2215 per_cpu_perf_limits 2216 Allow per-logical-CPU P-State performance control limits using 2217 cpufreq sysfs interface 2218 2219 intremap= [X86-64, Intel-IOMMU] 2220 on enable Interrupt Remapping (default) 2221 off disable Interrupt Remapping 2222 nosid disable Source ID checking 2223 no_x2apic_optout 2224 BIOS x2APIC opt-out request will be ignored 2225 nopost disable Interrupt Posting 2226 2227 iomem= Disable strict checking of access to MMIO memory 2228 strict regions from userspace. 2229 relaxed 2230 2231 iommu= [X86] 2232 off 2233 force 2234 noforce 2235 biomerge 2236 panic 2237 nopanic 2238 merge 2239 nomerge 2240 soft 2241 pt [X86] 2242 nopt [X86] 2243 nobypass [PPC/POWERNV] 2244 Disable IOMMU bypass, using IOMMU for PCI devices. 2245 2246 iommu.forcedac= [ARM64, X86] Control IOVA allocation for PCI devices. 2247 Format: { "0" | "1" } 2248 0 - Try to allocate a 32-bit DMA address first, before 2249 falling back to the full range if needed. 2250 1 - Allocate directly from the full usable range, 2251 forcing Dual Address Cycle for PCI cards supporting 2252 greater than 32-bit addressing. 2253 2254 iommu.strict= [ARM64, X86] Configure TLB invalidation behaviour 2255 Format: { "0" | "1" } 2256 0 - Lazy mode. 2257 Request that DMA unmap operations use deferred 2258 invalidation of hardware TLBs, for increased 2259 throughput at the cost of reduced device isolation. 2260 Will fall back to strict mode if not supported by 2261 the relevant IOMMU driver. 2262 1 - Strict mode. 2263 DMA unmap operations invalidate IOMMU hardware TLBs 2264 synchronously. 2265 unset - Use value of CONFIG_IOMMU_DEFAULT_DMA_{LAZY,STRICT}. 2266 Note: on x86, strict mode specified via one of the 2267 legacy driver-specific options takes precedence. 2268 2269 iommu.passthrough= 2270 [ARM64, X86] Configure DMA to bypass the IOMMU by default. 2271 Format: { "0" | "1" } 2272 0 - Use IOMMU translation for DMA. 2273 1 - Bypass the IOMMU for DMA. 2274 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH. 2275 2276 io7= [HW] IO7 for Marvel-based Alpha systems 2277 See comment before marvel_specify_io7 in 2278 arch/alpha/kernel/core_marvel.c. 2279 2280 io_delay= [X86] I/O delay method 2281 0x80 2282 Standard port 0x80 based delay 2283 0xed 2284 Alternate port 0xed based delay (needed on some systems) 2285 udelay 2286 Simple two microseconds delay 2287 none 2288 No delay 2289 2290 ip= [IP_PNP] 2291 See Documentation/admin-guide/nfs/nfsroot.rst. 2292 2293 ipcmni_extend [KNL] Extend the maximum number of unique System V 2294 IPC identifiers from 32,768 to 16,777,216. 2295 2296 irqaffinity= [SMP] Set the default irq affinity mask 2297 The argument is a cpu list, as described above. 2298 2299 irqchip.gicv2_force_probe= 2300 [ARM, ARM64] 2301 Format: <bool> 2302 Force the kernel to look for the second 4kB page 2303 of a GICv2 controller even if the memory range 2304 exposed by the device tree is too small. 2305 2306 irqchip.gicv3_nolpi= 2307 [ARM, ARM64] 2308 Force the kernel to ignore the availability of 2309 LPIs (and by consequence ITSs). Intended for system 2310 that use the kernel as a bootloader, and thus want 2311 to let secondary kernels in charge of setting up 2312 LPIs. 2313 2314 irqchip.gicv3_pseudo_nmi= [ARM64] 2315 Enables support for pseudo-NMIs in the kernel. This 2316 requires the kernel to be built with 2317 CONFIG_ARM64_PSEUDO_NMI. 2318 2319 irqfixup [HW] 2320 When an interrupt is not handled search all handlers 2321 for it. Intended to get systems with badly broken 2322 firmware running. 2323 2324 irqpoll [HW] 2325 When an interrupt is not handled search all handlers 2326 for it. Also check all handlers each timer 2327 interrupt. Intended to get systems with badly broken 2328 firmware running. 2329 2330 isapnp= [ISAPNP] 2331 Format: <RDP>,<reset>,<pci_scan>,<verbosity> 2332 2333 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance. 2334 [Deprecated - use cpusets instead] 2335 Format: [flag-list,]<cpu-list> 2336 2337 Specify one or more CPUs to isolate from disturbances 2338 specified in the flag list (default: domain): 2339 2340 nohz 2341 Disable the tick when a single task runs. 2342 2343 A residual 1Hz tick is offloaded to workqueues, which you 2344 need to affine to housekeeping through the global 2345 workqueue's affinity configured via the 2346 /sys/devices/virtual/workqueue/cpumask sysfs file, or 2347 by using the 'domain' flag described below. 2348 2349 NOTE: by default the global workqueue runs on all CPUs, 2350 so to protect individual CPUs the 'cpumask' file has to 2351 be configured manually after bootup. 2352 2353 domain 2354 Isolate from the general SMP balancing and scheduling 2355 algorithms. Note that performing domain isolation this way 2356 is irreversible: it's not possible to bring back a CPU to 2357 the domains once isolated through isolcpus. It's strongly 2358 advised to use cpusets instead to disable scheduler load 2359 balancing through the "cpuset.sched_load_balance" file. 2360 It offers a much more flexible interface where CPUs can 2361 move in and out of an isolated set anytime. 2362 2363 You can move a process onto or off an "isolated" CPU via 2364 the CPU affinity syscalls or cpuset. 2365 <cpu number> begins at 0 and the maximum value is 2366 "number of CPUs in system - 1". 2367 2368 managed_irq 2369 2370 Isolate from being targeted by managed interrupts 2371 which have an interrupt mask containing isolated 2372 CPUs. The affinity of managed interrupts is 2373 handled by the kernel and cannot be changed via 2374 the /proc/irq/* interfaces. 2375 2376 This isolation is best effort and only effective 2377 if the automatically assigned interrupt mask of a 2378 device queue contains isolated and housekeeping 2379 CPUs. If housekeeping CPUs are online then such 2380 interrupts are directed to the housekeeping CPU 2381 so that IO submitted on the housekeeping CPU 2382 cannot disturb the isolated CPU. 2383 2384 If a queue's affinity mask contains only isolated 2385 CPUs then this parameter has no effect on the 2386 interrupt routing decision, though interrupts are 2387 only delivered when tasks running on those 2388 isolated CPUs submit IO. IO submitted on 2389 housekeeping CPUs has no influence on those 2390 queues. 2391 2392 The format of <cpu-list> is described above. 2393 2394 iucv= [HW,NET] 2395 2396 ivrs_ioapic [HW,X86-64] 2397 Provide an override to the IOAPIC-ID<->DEVICE-ID 2398 mapping provided in the IVRS ACPI table. 2399 By default, PCI segment is 0, and can be omitted. 2400 2401 For example, to map IOAPIC-ID decimal 10 to 2402 PCI segment 0x1 and PCI device 00:14.0, 2403 write the parameter as: 2404 ivrs_ioapic=10@0001:00:14.0 2405 2406 Deprecated formats: 2407 * To map IOAPIC-ID decimal 10 to PCI device 00:14.0 2408 write the parameter as: 2409 ivrs_ioapic[10]=00:14.0 2410 * To map IOAPIC-ID decimal 10 to PCI segment 0x1 and 2411 PCI device 00:14.0 write the parameter as: 2412 ivrs_ioapic[10]=0001:00:14.0 2413 2414 ivrs_hpet [HW,X86-64] 2415 Provide an override to the HPET-ID<->DEVICE-ID 2416 mapping provided in the IVRS ACPI table. 2417 By default, PCI segment is 0, and can be omitted. 2418 2419 For example, to map HPET-ID decimal 10 to 2420 PCI segment 0x1 and PCI device 00:14.0, 2421 write the parameter as: 2422 ivrs_hpet=10@0001:00:14.0 2423 2424 Deprecated formats: 2425 * To map HPET-ID decimal 0 to PCI device 00:14.0 2426 write the parameter as: 2427 ivrs_hpet[0]=00:14.0 2428 * To map HPET-ID decimal 10 to PCI segment 0x1 and 2429 PCI device 00:14.0 write the parameter as: 2430 ivrs_ioapic[10]=0001:00:14.0 2431 2432 ivrs_acpihid [HW,X86-64] 2433 Provide an override to the ACPI-HID:UID<->DEVICE-ID 2434 mapping provided in the IVRS ACPI table. 2435 By default, PCI segment is 0, and can be omitted. 2436 2437 For example, to map UART-HID:UID AMD0020:0 to 2438 PCI segment 0x1 and PCI device ID 00:14.5, 2439 write the parameter as: 2440 ivrs_acpihid=AMD0020:0@0001:00:14.5 2441 2442 Deprecated formats: 2443 * To map UART-HID:UID AMD0020:0 to PCI segment is 0, 2444 PCI device ID 00:14.5, write the parameter as: 2445 ivrs_acpihid[00:14.5]=AMD0020:0 2446 * To map UART-HID:UID AMD0020:0 to PCI segment 0x1 and 2447 PCI device ID 00:14.5, write the parameter as: 2448 ivrs_acpihid[0001:00:14.5]=AMD0020:0 2449 2450 js= [HW,JOY] Analog joystick 2451 See Documentation/input/joydev/joystick.rst. 2452 2453 kasan_multi_shot 2454 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print 2455 report on every invalid memory access. Without this 2456 parameter KASAN will print report only for the first 2457 invalid access. 2458 2459 keep_bootcon [KNL] 2460 Do not unregister boot console at start. This is only 2461 useful for debugging when something happens in the window 2462 between unregistering the boot console and initializing 2463 the real console. 2464 2465 keepinitrd [HW,ARM] 2466 2467 kernelcore= [KNL,X86,IA-64,PPC] 2468 Format: nn[KMGTPE] | nn% | "mirror" 2469 This parameter specifies the amount of memory usable by 2470 the kernel for non-movable allocations. The requested 2471 amount is spread evenly throughout all nodes in the 2472 system as ZONE_NORMAL. The remaining memory is used for 2473 movable memory in its own zone, ZONE_MOVABLE. In the 2474 event, a node is too small to have both ZONE_NORMAL and 2475 ZONE_MOVABLE, kernelcore memory will take priority and 2476 other nodes will have a larger ZONE_MOVABLE. 2477 2478 ZONE_MOVABLE is used for the allocation of pages that 2479 may be reclaimed or moved by the page migration 2480 subsystem. Note that allocations like PTEs-from-HighMem 2481 still use the HighMem zone if it exists, and the Normal 2482 zone if it does not. 2483 2484 It is possible to specify the exact amount of memory in 2485 the form of "nn[KMGTPE]", a percentage of total system 2486 memory in the form of "nn%", or "mirror". If "mirror" 2487 option is specified, mirrored (reliable) memory is used 2488 for non-movable allocations and remaining memory is used 2489 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror" 2490 are exclusive, so you cannot specify multiple forms. 2491 2492 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port. 2493 Format: <Controller#>[,poll interval] 2494 The controller # is the number of the ehci usb debug 2495 port as it is probed via PCI. The poll interval is 2496 optional and is the number seconds in between 2497 each poll cycle to the debug port in case you need 2498 the functionality for interrupting the kernel with 2499 gdb or control-c on the dbgp connection. When 2500 not using this parameter you use sysrq-g to break into 2501 the kernel debugger. 2502 2503 kgdboc= [KGDB,HW] kgdb over consoles. 2504 Requires a tty driver that supports console polling, 2505 or a supported polling keyboard driver (non-usb). 2506 Serial only format: <serial_device>[,baud] 2507 keyboard only format: kbd 2508 keyboard and serial format: kbd,<serial_device>[,baud] 2509 Optional Kernel mode setting: 2510 kms, kbd format: kms,kbd 2511 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud] 2512 2513 kgdboc_earlycon= [KGDB,HW] 2514 If the boot console provides the ability to read 2515 characters and can work in polling mode, you can use 2516 this parameter to tell kgdb to use it as a backend 2517 until the normal console is registered. Intended to 2518 be used together with the kgdboc parameter which 2519 specifies the normal console to transition to. 2520 2521 The name of the early console should be specified 2522 as the value of this parameter. Note that the name of 2523 the early console might be different than the tty 2524 name passed to kgdboc. It's OK to leave the value 2525 blank and the first boot console that implements 2526 read() will be picked. 2527 2528 kgdbwait [KGDB] Stop kernel execution and enter the 2529 kernel debugger at the earliest opportunity. 2530 2531 kmac= [MIPS] Korina ethernet MAC address. 2532 Configure the RouterBoard 532 series on-chip 2533 Ethernet adapter MAC address. 2534 2535 kmemleak= [KNL] Boot-time kmemleak enable/disable 2536 Valid arguments: on, off 2537 Default: on 2538 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y, 2539 the default is off. 2540 2541 kprobe_event=[probe-list] 2542 [FTRACE] Add kprobe events and enable at boot time. 2543 The probe-list is a semicolon delimited list of probe 2544 definitions. Each definition is same as kprobe_events 2545 interface, but the parameters are comma delimited. 2546 For example, to add a kprobe event on vfs_read with 2547 arg1 and arg2, add to the command line; 2548 2549 kprobe_event=p,vfs_read,$arg1,$arg2 2550 2551 See also Documentation/trace/kprobetrace.rst "Kernel 2552 Boot Parameter" section. 2553 2554 kpti= [ARM64] Control page table isolation of user 2555 and kernel address spaces. 2556 Default: enabled on cores which need mitigation. 2557 0: force disabled 2558 1: force enabled 2559 2560 kunit.enable= [KUNIT] Enable executing KUnit tests. Requires 2561 CONFIG_KUNIT to be set to be fully enabled. The 2562 default value can be overridden via 2563 KUNIT_DEFAULT_ENABLED. 2564 Default is 1 (enabled) 2565 2566 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs. 2567 Default is 0 (don't ignore, but inject #GP) 2568 2569 kvm.eager_page_split= 2570 [KVM,X86] Controls whether or not KVM will try to 2571 proactively split all huge pages during dirty logging. 2572 Eager page splitting reduces interruptions to vCPU 2573 execution by eliminating the write-protection faults 2574 and MMU lock contention that would otherwise be 2575 required to split huge pages lazily. 2576 2577 VM workloads that rarely perform writes or that write 2578 only to a small region of VM memory may benefit from 2579 disabling eager page splitting to allow huge pages to 2580 still be used for reads. 2581 2582 The behavior of eager page splitting depends on whether 2583 KVM_DIRTY_LOG_INITIALLY_SET is enabled or disabled. If 2584 disabled, all huge pages in a memslot will be eagerly 2585 split when dirty logging is enabled on that memslot. If 2586 enabled, eager page splitting will be performed during 2587 the KVM_CLEAR_DIRTY ioctl, and only for the pages being 2588 cleared. 2589 2590 Eager page splitting is only supported when kvm.tdp_mmu=Y. 2591 2592 Default is Y (on). 2593 2594 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface. 2595 Default is false (don't support). 2596 2597 kvm.nx_huge_pages= 2598 [KVM] Controls the software workaround for the 2599 X86_BUG_ITLB_MULTIHIT bug. 2600 force : Always deploy workaround. 2601 off : Never deploy workaround. 2602 auto : Deploy workaround based on the presence of 2603 X86_BUG_ITLB_MULTIHIT. 2604 2605 Default is 'auto'. 2606 2607 If the software workaround is enabled for the host, 2608 guests do need not to enable it for nested guests. 2609 2610 kvm.nx_huge_pages_recovery_ratio= 2611 [KVM] Controls how many 4KiB pages are periodically zapped 2612 back to huge pages. 0 disables the recovery, otherwise if 2613 the value is N KVM will zap 1/Nth of the 4KiB pages every 2614 period (see below). The default is 60. 2615 2616 kvm.nx_huge_pages_recovery_period_ms= 2617 [KVM] Controls the time period at which KVM zaps 4KiB pages 2618 back to huge pages. If the value is a non-zero N, KVM will 2619 zap a portion (see ratio above) of the pages every N msecs. 2620 If the value is 0 (the default), KVM will pick a period based 2621 on the ratio, such that a page is zapped after 1 hour on average. 2622 2623 kvm-amd.nested= [KVM,AMD] Control nested virtualization feature in 2624 KVM/SVM. Default is 1 (enabled). 2625 2626 kvm-amd.npt= [KVM,AMD] Control KVM's use of Nested Page Tables, 2627 a.k.a. Two-Dimensional Page Tables. Default is 1 2628 (enabled). Disable by KVM if hardware lacks support 2629 for NPT. 2630 2631 kvm-arm.mode= 2632 [KVM,ARM] Select one of KVM/arm64's modes of operation. 2633 2634 none: Forcefully disable KVM. 2635 2636 nvhe: Standard nVHE-based mode, without support for 2637 protected guests. 2638 2639 protected: nVHE-based mode with support for guests whose 2640 state is kept private from the host. 2641 2642 nested: VHE-based mode with support for nested 2643 virtualization. Requires at least ARMv8.3 2644 hardware. 2645 2646 Defaults to VHE/nVHE based on hardware support. Setting 2647 mode to "protected" will disable kexec and hibernation 2648 for the host. "nested" is experimental and should be 2649 used with extreme caution. 2650 2651 kvm-arm.vgic_v3_group0_trap= 2652 [KVM,ARM] Trap guest accesses to GICv3 group-0 2653 system registers 2654 2655 kvm-arm.vgic_v3_group1_trap= 2656 [KVM,ARM] Trap guest accesses to GICv3 group-1 2657 system registers 2658 2659 kvm-arm.vgic_v3_common_trap= 2660 [KVM,ARM] Trap guest accesses to GICv3 common 2661 system registers 2662 2663 kvm-arm.vgic_v4_enable= 2664 [KVM,ARM] Allow use of GICv4 for direct injection of 2665 LPIs. 2666 2667 kvm_cma_resv_ratio=n [PPC] 2668 Reserves given percentage from system memory area for 2669 contiguous memory allocation for KVM hash pagetable 2670 allocation. 2671 By default it reserves 5% of total system memory. 2672 Format: <integer> 2673 Default: 5 2674 2675 kvm-intel.ept= [KVM,Intel] Control KVM's use of Extended Page Tables, 2676 a.k.a. Two-Dimensional Page Tables. Default is 1 2677 (enabled). Disable by KVM if hardware lacks support 2678 for EPT. 2679 2680 kvm-intel.emulate_invalid_guest_state= 2681 [KVM,Intel] Control whether to emulate invalid guest 2682 state. Ignored if kvm-intel.enable_unrestricted_guest=1, 2683 as guest state is never invalid for unrestricted 2684 guests. This param doesn't apply to nested guests (L2), 2685 as KVM never emulates invalid L2 guest state. 2686 Default is 1 (enabled). 2687 2688 kvm-intel.flexpriority= 2689 [KVM,Intel] Control KVM's use of FlexPriority feature 2690 (TPR shadow). Default is 1 (enabled). Disable by KVM if 2691 hardware lacks support for it. 2692 2693 kvm-intel.nested= 2694 [KVM,Intel] Control nested virtualization feature in 2695 KVM/VMX. Default is 1 (enabled). 2696 2697 kvm-intel.unrestricted_guest= 2698 [KVM,Intel] Control KVM's use of unrestricted guest 2699 feature (virtualized real and unpaged mode). Default 2700 is 1 (enabled). Disable by KVM if EPT is disabled or 2701 hardware lacks support for it. 2702 2703 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault 2704 CVE-2018-3620. 2705 2706 Valid arguments: never, cond, always 2707 2708 always: L1D cache flush on every VMENTER. 2709 cond: Flush L1D on VMENTER only when the code between 2710 VMEXIT and VMENTER can leak host memory. 2711 never: Disables the mitigation 2712 2713 Default is cond (do L1 cache flush in specific instances) 2714 2715 kvm-intel.vpid= [KVM,Intel] Control KVM's use of Virtual Processor 2716 Identification feature (tagged TLBs). Default is 1 2717 (enabled). Disable by KVM if hardware lacks support 2718 for it. 2719 2720 l1d_flush= [X86,INTEL] 2721 Control mitigation for L1D based snooping vulnerability. 2722 2723 Certain CPUs are vulnerable to an exploit against CPU 2724 internal buffers which can forward information to a 2725 disclosure gadget under certain conditions. 2726 2727 In vulnerable processors, the speculatively 2728 forwarded data can be used in a cache side channel 2729 attack, to access data to which the attacker does 2730 not have direct access. 2731 2732 This parameter controls the mitigation. The 2733 options are: 2734 2735 on - enable the interface for the mitigation 2736 2737 l1tf= [X86] Control mitigation of the L1TF vulnerability on 2738 affected CPUs 2739 2740 The kernel PTE inversion protection is unconditionally 2741 enabled and cannot be disabled. 2742 2743 full 2744 Provides all available mitigations for the 2745 L1TF vulnerability. Disables SMT and 2746 enables all mitigations in the 2747 hypervisors, i.e. unconditional L1D flush. 2748 2749 SMT control and L1D flush control via the 2750 sysfs interface is still possible after 2751 boot. Hypervisors will issue a warning 2752 when the first VM is started in a 2753 potentially insecure configuration, 2754 i.e. SMT enabled or L1D flush disabled. 2755 2756 full,force 2757 Same as 'full', but disables SMT and L1D 2758 flush runtime control. Implies the 2759 'nosmt=force' command line option. 2760 (i.e. sysfs control of SMT is disabled.) 2761 2762 flush 2763 Leaves SMT enabled and enables the default 2764 hypervisor mitigation, i.e. conditional 2765 L1D flush. 2766 2767 SMT control and L1D flush control via the 2768 sysfs interface is still possible after 2769 boot. Hypervisors will issue a warning 2770 when the first VM is started in a 2771 potentially insecure configuration, 2772 i.e. SMT enabled or L1D flush disabled. 2773 2774 flush,nosmt 2775 2776 Disables SMT and enables the default 2777 hypervisor mitigation. 2778 2779 SMT control and L1D flush control via the 2780 sysfs interface is still possible after 2781 boot. Hypervisors will issue a warning 2782 when the first VM is started in a 2783 potentially insecure configuration, 2784 i.e. SMT enabled or L1D flush disabled. 2785 2786 flush,nowarn 2787 Same as 'flush', but hypervisors will not 2788 warn when a VM is started in a potentially 2789 insecure configuration. 2790 2791 off 2792 Disables hypervisor mitigations and doesn't 2793 emit any warnings. 2794 It also drops the swap size and available 2795 RAM limit restriction on both hypervisor and 2796 bare metal. 2797 2798 Default is 'flush'. 2799 2800 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst 2801 2802 l2cr= [PPC] 2803 2804 l3cr= [PPC] 2805 2806 lapic [X86-32,APIC] Enable the local APIC even if BIOS 2807 disabled it. 2808 2809 lapic= [X86,APIC] Do not use TSC deadline 2810 value for LAPIC timer one-shot implementation. Default 2811 back to the programmable timer unit in the LAPIC. 2812 Format: notscdeadline 2813 2814 lapic_timer_c2_ok [X86,APIC] trust the local apic timer 2815 in C2 power state. 2816 2817 libata.dma= [LIBATA] DMA control 2818 libata.dma=0 Disable all PATA and SATA DMA 2819 libata.dma=1 PATA and SATA Disk DMA only 2820 libata.dma=2 ATAPI (CDROM) DMA only 2821 libata.dma=4 Compact Flash DMA only 2822 Combinations also work, so libata.dma=3 enables DMA 2823 for disks and CDROMs, but not CFs. 2824 2825 libata.ignore_hpa= [LIBATA] Ignore HPA limit 2826 libata.ignore_hpa=0 keep BIOS limits (default) 2827 libata.ignore_hpa=1 ignore limits, using full disk 2828 2829 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume 2830 when set. 2831 Format: <int> 2832 2833 libata.force= [LIBATA] Force configurations. The format is a comma- 2834 separated list of "[ID:]VAL" where ID is PORT[.DEVICE]. 2835 PORT and DEVICE are decimal numbers matching port, link 2836 or device. Basically, it matches the ATA ID string 2837 printed on console by libata. If the whole ID part is 2838 omitted, the last PORT and DEVICE values are used. If 2839 ID hasn't been specified yet, the configuration applies 2840 to all ports, links and devices. 2841 2842 If only DEVICE is omitted, the parameter applies to 2843 the port and all links and devices behind it. DEVICE 2844 number of 0 either selects the first device or the 2845 first fan-out link behind PMP device. It does not 2846 select the host link. DEVICE number of 15 selects the 2847 host link and device attached to it. 2848 2849 The VAL specifies the configuration to force. As long 2850 as there is no ambiguity, shortcut notation is allowed. 2851 For example, both 1.5 and 1.5G would work for 1.5Gbps. 2852 The following configurations can be forced. 2853 2854 * Cable type: 40c, 80c, short40c, unk, ign or sata. 2855 Any ID with matching PORT is used. 2856 2857 * SATA link speed limit: 1.5Gbps or 3.0Gbps. 2858 2859 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7]. 2860 udma[/][16,25,33,44,66,100,133] notation is also 2861 allowed. 2862 2863 * nohrst, nosrst, norst: suppress hard, soft and both 2864 resets. 2865 2866 * rstonce: only attempt one reset during hot-unplug 2867 link recovery. 2868 2869 * [no]dbdelay: Enable or disable the extra 200ms delay 2870 before debouncing a link PHY and device presence 2871 detection. 2872 2873 * [no]ncq: Turn on or off NCQ. 2874 2875 * [no]ncqtrim: Enable or disable queued DSM TRIM. 2876 2877 * [no]ncqati: Enable or disable NCQ trim on ATI chipset. 2878 2879 * [no]trim: Enable or disable (unqueued) TRIM. 2880 2881 * trim_zero: Indicate that TRIM command zeroes data. 2882 2883 * max_trim_128m: Set 128M maximum trim size limit. 2884 2885 * [no]dma: Turn on or off DMA transfers. 2886 2887 * atapi_dmadir: Enable ATAPI DMADIR bridge support. 2888 2889 * atapi_mod16_dma: Enable the use of ATAPI DMA for 2890 commands that are not a multiple of 16 bytes. 2891 2892 * [no]dmalog: Enable or disable the use of the 2893 READ LOG DMA EXT command to access logs. 2894 2895 * [no]iddevlog: Enable or disable access to the 2896 identify device data log. 2897 2898 * [no]logdir: Enable or disable access to the general 2899 purpose log directory. 2900 2901 * max_sec_128: Set transfer size limit to 128 sectors. 2902 2903 * max_sec_1024: Set or clear transfer size limit to 2904 1024 sectors. 2905 2906 * max_sec_lba48: Set or clear transfer size limit to 2907 65535 sectors. 2908 2909 * [no]lpm: Enable or disable link power management. 2910 2911 * [no]setxfer: Indicate if transfer speed mode setting 2912 should be skipped. 2913 2914 * [no]fua: Disable or enable FUA (Force Unit Access) 2915 support for devices supporting this feature. 2916 2917 * dump_id: Dump IDENTIFY data. 2918 2919 * disable: Disable this device. 2920 2921 If there are multiple matching configurations changing 2922 the same attribute, the last one is used. 2923 2924 load_ramdisk= [RAM] [Deprecated] 2925 2926 lockd.nlm_grace_period=P [NFS] Assign grace period. 2927 Format: <integer> 2928 2929 lockd.nlm_tcpport=N [NFS] Assign TCP port. 2930 Format: <integer> 2931 2932 lockd.nlm_timeout=T [NFS] Assign timeout value. 2933 Format: <integer> 2934 2935 lockd.nlm_udpport=M [NFS] Assign UDP port. 2936 Format: <integer> 2937 2938 lockdown= [SECURITY] 2939 { integrity | confidentiality } 2940 Enable the kernel lockdown feature. If set to 2941 integrity, kernel features that allow userland to 2942 modify the running kernel are disabled. If set to 2943 confidentiality, kernel features that allow userland 2944 to extract confidential information from the kernel 2945 are also disabled. 2946 2947 locktorture.nreaders_stress= [KNL] 2948 Set the number of locking read-acquisition kthreads. 2949 Defaults to being automatically set based on the 2950 number of online CPUs. 2951 2952 locktorture.nwriters_stress= [KNL] 2953 Set the number of locking write-acquisition kthreads. 2954 2955 locktorture.onoff_holdoff= [KNL] 2956 Set time (s) after boot for CPU-hotplug testing. 2957 2958 locktorture.onoff_interval= [KNL] 2959 Set time (s) between CPU-hotplug operations, or 2960 zero to disable CPU-hotplug testing. 2961 2962 locktorture.shuffle_interval= [KNL] 2963 Set task-shuffle interval (jiffies). Shuffling 2964 tasks allows some CPUs to go into dyntick-idle 2965 mode during the locktorture test. 2966 2967 locktorture.shutdown_secs= [KNL] 2968 Set time (s) after boot system shutdown. This 2969 is useful for hands-off automated testing. 2970 2971 locktorture.stat_interval= [KNL] 2972 Time (s) between statistics printk()s. 2973 2974 locktorture.stutter= [KNL] 2975 Time (s) to stutter testing, for example, 2976 specifying five seconds causes the test to run for 2977 five seconds, wait for five seconds, and so on. 2978 This tests the locking primitive's ability to 2979 transition abruptly to and from idle. 2980 2981 locktorture.torture_type= [KNL] 2982 Specify the locking implementation to test. 2983 2984 locktorture.writer_fifo= [KNL] 2985 Run the write-side locktorture kthreads at 2986 sched_set_fifo() real-time priority. 2987 2988 locktorture.verbose= [KNL] 2989 Enable additional printk() statements. 2990 2991 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver 2992 Format: <irq> 2993 2994 loglevel= All Kernel Messages with a loglevel smaller than the 2995 console loglevel will be printed to the console. It can 2996 also be changed with klogd or other programs. The 2997 loglevels are defined as follows: 2998 2999 0 (KERN_EMERG) system is unusable 3000 1 (KERN_ALERT) action must be taken immediately 3001 2 (KERN_CRIT) critical conditions 3002 3 (KERN_ERR) error conditions 3003 4 (KERN_WARNING) warning conditions 3004 5 (KERN_NOTICE) normal but significant condition 3005 6 (KERN_INFO) informational 3006 7 (KERN_DEBUG) debug-level messages 3007 3008 log_buf_len=n[KMG] Sets the size of the printk ring buffer, 3009 in bytes. n must be a power of two and greater 3010 than the minimal size. The minimal size is defined 3011 by LOG_BUF_SHIFT kernel config parameter. There is 3012 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter 3013 that allows to increase the default size depending on 3014 the number of CPUs. See init/Kconfig for more details. 3015 3016 logo.nologo [FB] Disables display of the built-in Linux logo. 3017 This may be used to provide more screen space for 3018 kernel log messages and is useful when debugging 3019 kernel boot problems. 3020 3021 lp=0 [LP] Specify parallel ports to use, e.g, 3022 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses 3023 lp=reset first parallel port). 'lp=0' disables the 3024 lp=auto printer driver. 'lp=reset' (which can be 3025 specified in addition to the ports) causes 3026 attached printers to be reset. Using 3027 lp=port1,port2,... specifies the parallel ports 3028 to associate lp devices with, starting with 3029 lp0. A port specification may be 'none' to skip 3030 that lp device, or a parport name such as 3031 'parport0'. Specifying 'lp=auto' instead of a 3032 port specification list means that device IDs 3033 from each port should be examined, to see if 3034 an IEEE 1284-compliant printer is attached; if 3035 so, the driver will manage that printer. 3036 See also header of drivers/char/lp.c. 3037 3038 lpj=n [KNL] 3039 Sets loops_per_jiffy to given constant, thus avoiding 3040 time-consuming boot-time autodetection (up to 250 ms per 3041 CPU). 0 enables autodetection (default). To determine 3042 the correct value for your kernel, boot with normal 3043 autodetection and see what value is printed. Note that 3044 on SMP systems the preset will be applied to all CPUs, 3045 which is likely to cause problems if your CPUs need 3046 significantly divergent settings. An incorrect value 3047 will cause delays in the kernel to be wrong, leading to 3048 unpredictable I/O errors and other breakage. Although 3049 unlikely, in the extreme case this might damage your 3050 hardware. 3051 3052 ltpc= [NET] 3053 Format: <io>,<irq>,<dma> 3054 3055 lsm.debug [SECURITY] Enable LSM initialization debugging output. 3056 3057 lsm=lsm1,...,lsmN 3058 [SECURITY] Choose order of LSM initialization. This 3059 overrides CONFIG_LSM, and the "security=" parameter. 3060 3061 machvec= [IA-64] Force the use of a particular machine-vector 3062 (machvec) in a generic kernel. 3063 Example: machvec=hpzx1 3064 3065 machtype= [Loongson] Share the same kernel image file between 3066 different yeeloong laptops. 3067 Example: machtype=lemote-yeeloong-2f-7inch 3068 3069 max_addr=nn[KMG] [KNL,BOOT,IA-64] All physical memory greater 3070 than or equal to this physical address is ignored. 3071 3072 maxcpus= [SMP] Maximum number of processors that an SMP kernel 3073 will bring up during bootup. maxcpus=n : n >= 0 limits 3074 the kernel to bring up 'n' processors. Surely after 3075 bootup you can bring up the other plugged cpu by executing 3076 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus 3077 only takes effect during system bootup. 3078 While n=0 is a special case, it is equivalent to "nosmp", 3079 which also disables the IO APIC. 3080 3081 max_loop= [LOOP] The number of loop block devices that get 3082 (loop.max_loop) unconditionally pre-created at init time. The default 3083 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead 3084 of statically allocating a predefined number, loop 3085 devices can be requested on-demand with the 3086 /dev/loop-control interface. 3087 3088 mce [X86-32] Machine Check Exception 3089 3090 mce=option [X86-64] See Documentation/arch/x86/x86_64/boot-options.rst 3091 3092 md= [HW] RAID subsystems devices and level 3093 See Documentation/admin-guide/md.rst. 3094 3095 mdacon= [MDA] 3096 Format: <first>,<last> 3097 Specifies range of consoles to be captured by the MDA. 3098 3099 mds= [X86,INTEL] 3100 Control mitigation for the Micro-architectural Data 3101 Sampling (MDS) vulnerability. 3102 3103 Certain CPUs are vulnerable to an exploit against CPU 3104 internal buffers which can forward information to a 3105 disclosure gadget under certain conditions. 3106 3107 In vulnerable processors, the speculatively 3108 forwarded data can be used in a cache side channel 3109 attack, to access data to which the attacker does 3110 not have direct access. 3111 3112 This parameter controls the MDS mitigation. The 3113 options are: 3114 3115 full - Enable MDS mitigation on vulnerable CPUs 3116 full,nosmt - Enable MDS mitigation and disable 3117 SMT on vulnerable CPUs 3118 off - Unconditionally disable MDS mitigation 3119 3120 On TAA-affected machines, mds=off can be prevented by 3121 an active TAA mitigation as both vulnerabilities are 3122 mitigated with the same mechanism so in order to disable 3123 this mitigation, you need to specify tsx_async_abort=off 3124 too. 3125 3126 Not specifying this option is equivalent to 3127 mds=full. 3128 3129 For details see: Documentation/admin-guide/hw-vuln/mds.rst 3130 3131 mem=nn[KMG] [HEXAGON] Set the memory size. 3132 Must be specified, otherwise memory size will be 0. 3133 3134 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory 3135 Amount of memory to be used in cases as follows: 3136 3137 1 for test; 3138 2 when the kernel is not able to see the whole system memory; 3139 3 memory that lies after 'mem=' boundary is excluded from 3140 the hypervisor, then assigned to KVM guests. 3141 4 to limit the memory available for kdump kernel. 3142 3143 [ARC,MICROBLAZE] - the limit applies only to low memory, 3144 high memory is not affected. 3145 3146 [ARM64] - only limits memory covered by the linear 3147 mapping. The NOMAP regions are not affected. 3148 3149 [X86] Work as limiting max address. Use together 3150 with memmap= to avoid physical address space collisions. 3151 Without memmap= PCI devices could be placed at addresses 3152 belonging to unused RAM. 3153 3154 Note that this only takes effects during boot time since 3155 in above case 3, memory may need be hot added after boot 3156 if system memory of hypervisor is not sufficient. 3157 3158 mem=nn[KMG]@ss[KMG] 3159 [ARM,MIPS] - override the memory layout reported by 3160 firmware. 3161 Define a memory region of size nn[KMG] starting at 3162 ss[KMG]. 3163 Multiple different regions can be specified with 3164 multiple mem= parameters on the command line. 3165 3166 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel 3167 memory. 3168 3169 memblock=debug [KNL] Enable memblock debug messages. 3170 3171 memchunk=nn[KMG] 3172 [KNL,SH] Allow user to override the default size for 3173 per-device physically contiguous DMA buffers. 3174 3175 memhp_default_state=online/offline/online_kernel/online_movable 3176 [KNL] Set the initial state for the memory hotplug 3177 onlining policy. If not specified, the default value is 3178 set according to the 3179 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config 3180 option. 3181 See Documentation/admin-guide/mm/memory-hotplug.rst. 3182 3183 memmap=exactmap [KNL,X86] Enable setting of an exact 3184 E820 memory map, as specified by the user. 3185 Such memmap=exactmap lines can be constructed based on 3186 BIOS output or other requirements. See the memmap=nn@ss 3187 option description. 3188 3189 memmap=nn[KMG]@ss[KMG] 3190 [KNL, X86, MIPS, XTENSA] Force usage of a specific region of memory. 3191 Region of memory to be used is from ss to ss+nn. 3192 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG], 3193 which limits max address to nn[KMG]. 3194 Multiple different regions can be specified, 3195 comma delimited. 3196 Example: 3197 memmap=100M@2G,100M#3G,1G!1024G 3198 3199 memmap=nn[KMG]#ss[KMG] 3200 [KNL,ACPI] Mark specific memory as ACPI data. 3201 Region of memory to be marked is from ss to ss+nn. 3202 3203 memmap=nn[KMG]$ss[KMG] 3204 [KNL,ACPI] Mark specific memory as reserved. 3205 Region of memory to be reserved is from ss to ss+nn. 3206 Example: Exclude memory from 0x18690000-0x1869ffff 3207 memmap=64K$0x18690000 3208 or 3209 memmap=0x10000$0x18690000 3210 Some bootloaders may need an escape character before '$', 3211 like Grub2, otherwise '$' and the following number 3212 will be eaten. 3213 3214 memmap=nn[KMG]!ss[KMG] 3215 [KNL,X86] Mark specific memory as protected. 3216 Region of memory to be used, from ss to ss+nn. 3217 The memory region may be marked as e820 type 12 (0xc) 3218 and is NVDIMM or ADR memory. 3219 3220 memmap=<size>%<offset>-<oldtype>+<newtype> 3221 [KNL,ACPI] Convert memory within the specified region 3222 from <oldtype> to <newtype>. If "-<oldtype>" is left 3223 out, the whole region will be marked as <newtype>, 3224 even if previously unavailable. If "+<newtype>" is left 3225 out, matching memory will be removed. Types are 3226 specified as e820 types, e.g., 1 = RAM, 2 = reserved, 3227 3 = ACPI, 12 = PRAM. 3228 3229 memory_corruption_check=0/1 [X86] 3230 Some BIOSes seem to corrupt the first 64k of 3231 memory when doing things like suspend/resume. 3232 Setting this option will scan the memory 3233 looking for corruption. Enabling this will 3234 both detect corruption and prevent the kernel 3235 from using the memory being corrupted. 3236 However, its intended as a diagnostic tool; if 3237 repeatable BIOS-originated corruption always 3238 affects the same memory, you can use memmap= 3239 to prevent the kernel from using that memory. 3240 3241 memory_corruption_check_size=size [X86] 3242 By default it checks for corruption in the low 3243 64k, making this memory unavailable for normal 3244 use. Use this parameter to scan for 3245 corruption in more or less memory. 3246 3247 memory_corruption_check_period=seconds [X86] 3248 By default it checks for corruption every 60 3249 seconds. Use this parameter to check at some 3250 other rate. 0 disables periodic checking. 3251 3252 memory_hotplug.memmap_on_memory 3253 [KNL,X86,ARM] Boolean flag to enable this feature. 3254 Format: {on | off (default)} 3255 When enabled, runtime hotplugged memory will 3256 allocate its internal metadata (struct pages, 3257 those vmemmap pages cannot be optimized even 3258 if hugetlb_free_vmemmap is enabled) from the 3259 hotadded memory which will allow to hotadd a 3260 lot of memory without requiring additional 3261 memory to do so. 3262 This feature is disabled by default because it 3263 has some implication on large (e.g. GB) 3264 allocations in some configurations (e.g. small 3265 memory blocks). 3266 The state of the flag can be read in 3267 /sys/module/memory_hotplug/parameters/memmap_on_memory. 3268 Note that even when enabled, there are a few cases where 3269 the feature is not effective. 3270 3271 memtest= [KNL,X86,ARM,M68K,PPC,RISCV] Enable memtest 3272 Format: <integer> 3273 default : 0 <disable> 3274 Specifies the number of memtest passes to be 3275 performed. Each pass selects another test 3276 pattern from a given set of patterns. Memtest 3277 fills the memory with this pattern, validates 3278 memory contents and reserves bad memory 3279 regions that are detected. 3280 3281 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control 3282 Valid arguments: on, off 3283 Default: off 3284 mem_encrypt=on: Activate SME 3285 mem_encrypt=off: Do not activate SME 3286 3287 Refer to Documentation/virt/kvm/x86/amd-memory-encryption.rst 3288 for details on when memory encryption can be activated. 3289 3290 mem_sleep_default= [SUSPEND] Default system suspend mode: 3291 s2idle - Suspend-To-Idle 3292 shallow - Power-On Suspend or equivalent (if supported) 3293 deep - Suspend-To-RAM or equivalent (if supported) 3294 See Documentation/admin-guide/pm/sleep-states.rst. 3295 3296 mfgpt_irq= [IA-32] Specify the IRQ to use for the 3297 Multi-Function General Purpose Timers on AMD Geode 3298 platforms. 3299 3300 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when 3301 the BIOS has incorrectly applied a workaround. TinyBIOS 3302 version 0.98 is known to be affected, 0.99 fixes the 3303 problem by letting the user disable the workaround. 3304 3305 mga= [HW,DRM] 3306 3307 microcode.force_minrev= [X86] 3308 Format: <bool> 3309 Enable or disable the microcode minimal revision 3310 enforcement for the runtime microcode loader. 3311 3312 min_addr=nn[KMG] [KNL,BOOT,IA-64] All physical memory below this 3313 physical address is ignored. 3314 3315 mini2440= [ARM,HW,KNL] 3316 Format:[0..2][b][c][t] 3317 Default: "0tb" 3318 MINI2440 configuration specification: 3319 0 - The attached screen is the 3.5" TFT 3320 1 - The attached screen is the 7" TFT 3321 2 - The VGA Shield is attached (1024x768) 3322 Leaving out the screen size parameter will not load 3323 the TFT driver, and the framebuffer will be left 3324 unconfigured. 3325 b - Enable backlight. The TFT backlight pin will be 3326 linked to the kernel VESA blanking code and a GPIO 3327 LED. This parameter is not necessary when using the 3328 VGA shield. 3329 c - Enable the s3c camera interface. 3330 t - Reserved for enabling touchscreen support. The 3331 touchscreen support is not enabled in the mainstream 3332 kernel as of 2.6.30, a preliminary port can be found 3333 in the "bleeding edge" mini2440 support kernel at 3334 https://repo.or.cz/w/linux-2.6/mini2440.git 3335 3336 mitigations= 3337 [X86,PPC,S390,ARM64] Control optional mitigations for 3338 CPU vulnerabilities. This is a set of curated, 3339 arch-independent options, each of which is an 3340 aggregation of existing arch-specific options. 3341 3342 Note, "mitigations" is supported if and only if the 3343 kernel was built with CPU_MITIGATIONS=y. 3344 3345 off 3346 Disable all optional CPU mitigations. This 3347 improves system performance, but it may also 3348 expose users to several CPU vulnerabilities. 3349 Equivalent to: if nokaslr then kpti=0 [ARM64] 3350 gather_data_sampling=off [X86] 3351 indirect_target_selection=off [X86] 3352 kvm.nx_huge_pages=off [X86] 3353 l1tf=off [X86] 3354 mds=off [X86] 3355 mmio_stale_data=off [X86] 3356 no_entry_flush [PPC] 3357 no_uaccess_flush [PPC] 3358 nobp=0 [S390] 3359 nopti [X86,PPC] 3360 nospectre_bhb [ARM64] 3361 nospectre_v1 [X86,PPC] 3362 nospectre_v2 [X86,PPC,S390,ARM64] 3363 reg_file_data_sampling=off [X86] 3364 retbleed=off [X86] 3365 spec_store_bypass_disable=off [X86,PPC] 3366 spectre_bhi=off [X86] 3367 spectre_v2_user=off [X86] 3368 srbds=off [X86,INTEL] 3369 ssbd=force-off [ARM64] 3370 tsx_async_abort=off [X86] 3371 3372 Exceptions: 3373 This does not have any effect on 3374 kvm.nx_huge_pages when 3375 kvm.nx_huge_pages=force. 3376 3377 auto (default) 3378 Mitigate all CPU vulnerabilities, but leave SMT 3379 enabled, even if it's vulnerable. This is for 3380 users who don't want to be surprised by SMT 3381 getting disabled across kernel upgrades, or who 3382 have other ways of avoiding SMT-based attacks. 3383 Equivalent to: (default behavior) 3384 3385 auto,nosmt 3386 Mitigate all CPU vulnerabilities, disabling SMT 3387 if needed. This is for users who always want to 3388 be fully mitigated, even if it means losing SMT. 3389 Equivalent to: l1tf=flush,nosmt [X86] 3390 mds=full,nosmt [X86] 3391 tsx_async_abort=full,nosmt [X86] 3392 mmio_stale_data=full,nosmt [X86] 3393 retbleed=auto,nosmt [X86] 3394 3395 mminit_loglevel= 3396 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this 3397 parameter allows control of the logging verbosity for 3398 the additional memory initialisation checks. A value 3399 of 0 disables mminit logging and a level of 4 will 3400 log everything. Information is printed at KERN_DEBUG 3401 so loglevel=8 may also need to be specified. 3402 3403 mmio_stale_data= 3404 [X86,INTEL] Control mitigation for the Processor 3405 MMIO Stale Data vulnerabilities. 3406 3407 Processor MMIO Stale Data is a class of 3408 vulnerabilities that may expose data after an MMIO 3409 operation. Exposed data could originate or end in 3410 the same CPU buffers as affected by MDS and TAA. 3411 Therefore, similar to MDS and TAA, the mitigation 3412 is to clear the affected CPU buffers. 3413 3414 This parameter controls the mitigation. The 3415 options are: 3416 3417 full - Enable mitigation on vulnerable CPUs 3418 3419 full,nosmt - Enable mitigation and disable SMT on 3420 vulnerable CPUs. 3421 3422 off - Unconditionally disable mitigation 3423 3424 On MDS or TAA affected machines, 3425 mmio_stale_data=off can be prevented by an active 3426 MDS or TAA mitigation as these vulnerabilities are 3427 mitigated with the same mechanism so in order to 3428 disable this mitigation, you need to specify 3429 mds=off and tsx_async_abort=off too. 3430 3431 Not specifying this option is equivalent to 3432 mmio_stale_data=full. 3433 3434 For details see: 3435 Documentation/admin-guide/hw-vuln/processor_mmio_stale_data.rst 3436 3437 <module>.async_probe[=<bool>] [KNL] 3438 If no <bool> value is specified or if the value 3439 specified is not a valid <bool>, enable asynchronous 3440 probe on this module. Otherwise, enable/disable 3441 asynchronous probe on this module as indicated by the 3442 <bool> value. See also: module.async_probe 3443 3444 module.async_probe=<bool> 3445 [KNL] When set to true, modules will use async probing 3446 by default. To enable/disable async probing for a 3447 specific module, use the module specific control that 3448 is documented under <module>.async_probe. When both 3449 module.async_probe and <module>.async_probe are 3450 specified, <module>.async_probe takes precedence for 3451 the specific module. 3452 3453 module.enable_dups_trace 3454 [KNL] When CONFIG_MODULE_DEBUG_AUTOLOAD_DUPS is set, 3455 this means that duplicate request_module() calls will 3456 trigger a WARN_ON() instead of a pr_warn(). Note that 3457 if MODULE_DEBUG_AUTOLOAD_DUPS_TRACE is set, WARN_ON()s 3458 will always be issued and this option does nothing. 3459 module.sig_enforce 3460 [KNL] When CONFIG_MODULE_SIG is set, this means that 3461 modules without (valid) signatures will fail to load. 3462 Note that if CONFIG_MODULE_SIG_FORCE is set, that 3463 is always true, so this option does nothing. 3464 3465 module_blacklist= [KNL] Do not load a comma-separated list of 3466 modules. Useful for debugging problem modules. 3467 3468 mousedev.tap_time= 3469 [MOUSE] Maximum time between finger touching and 3470 leaving touchpad surface for touch to be considered 3471 a tap and be reported as a left button click (for 3472 touchpads working in absolute mode only). 3473 Format: <msecs> 3474 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices 3475 reporting absolute coordinates, such as tablets 3476 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices 3477 reporting absolute coordinates, such as tablets 3478 3479 movablecore= [KNL,X86,IA-64,PPC] 3480 Format: nn[KMGTPE] | nn% 3481 This parameter is the complement to kernelcore=, it 3482 specifies the amount of memory used for migratable 3483 allocations. If both kernelcore and movablecore is 3484 specified, then kernelcore will be at *least* the 3485 specified value but may be more. If movablecore on its 3486 own is specified, the administrator must be careful 3487 that the amount of memory usable for all allocations 3488 is not too small. 3489 3490 movable_node [KNL] Boot-time switch to make hotplugable memory 3491 NUMA nodes to be movable. This means that the memory 3492 of such nodes will be usable only for movable 3493 allocations which rules out almost all kernel 3494 allocations. Use with caution! 3495 3496 MTD_Partition= [MTD] 3497 Format: <name>,<region-number>,<size>,<offset> 3498 3499 MTD_Region= [MTD] Format: 3500 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>] 3501 3502 mtdparts= [MTD] 3503 See drivers/mtd/parsers/cmdlinepart.c 3504 3505 mtdset= [ARM] 3506 ARM/S3C2412 JIVE boot control 3507 3508 See arch/arm/mach-s3c/mach-jive.c 3509 3510 mtouchusb.raw_coordinates= 3511 [HW] Make the MicroTouch USB driver use raw coordinates 3512 ('y', default) or cooked coordinates ('n') 3513 3514 mtrr=debug [X86] 3515 Enable printing debug information related to MTRR 3516 registers at boot time. 3517 3518 mtrr_chunk_size=nn[KMG] [X86] 3519 used for mtrr cleanup. It is largest continuous chunk 3520 that could hold holes aka. UC entries. 3521 3522 mtrr_gran_size=nn[KMG] [X86] 3523 Used for mtrr cleanup. It is granularity of mtrr block. 3524 Default is 1. 3525 Large value could prevent small alignment from 3526 using up MTRRs. 3527 3528 mtrr_spare_reg_nr=n [X86] 3529 Format: <integer> 3530 Range: 0,7 : spare reg number 3531 Default : 1 3532 Used for mtrr cleanup. It is spare mtrr entries number. 3533 Set to 2 or more if your graphical card needs more. 3534 3535 multitce=off [PPC] This parameter disables the use of the pSeries 3536 firmware feature for updating multiple TCE entries 3537 at a time. 3538 3539 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card 3540 3541 netdev= [NET] Network devices parameters 3542 Format: <irq>,<io>,<mem_start>,<mem_end>,<name> 3543 Note that mem_start is often overloaded to mean 3544 something different and driver-specific. 3545 This usage is only documented in each driver source 3546 file if at all. 3547 3548 netpoll.carrier_timeout= 3549 [NET] Specifies amount of time (in seconds) that 3550 netpoll should wait for a carrier. By default netpoll 3551 waits 4 seconds. 3552 3553 nf_conntrack.acct= 3554 [NETFILTER] Enable connection tracking flow accounting 3555 0 to disable accounting 3556 1 to enable accounting 3557 Default value is 0. 3558 3559 nfs.cache_getent= 3560 [NFS] sets the pathname to the program which is used 3561 to update the NFS client cache entries. 3562 3563 nfs.cache_getent_timeout= 3564 [NFS] sets the timeout after which an attempt to 3565 update a cache entry is deemed to have failed. 3566 3567 nfs.callback_nr_threads= 3568 [NFSv4] set the total number of threads that the 3569 NFS client will assign to service NFSv4 callback 3570 requests. 3571 3572 nfs.callback_tcpport= 3573 [NFS] set the TCP port on which the NFSv4 callback 3574 channel should listen. 3575 3576 nfs.enable_ino64= 3577 [NFS] enable 64-bit inode numbers. 3578 If zero, the NFS client will fake up a 32-bit inode 3579 number for the readdir() and stat() syscalls instead 3580 of returning the full 64-bit number. 3581 The default is to return 64-bit inode numbers. 3582 3583 nfs.idmap_cache_timeout= 3584 [NFS] set the maximum lifetime for idmapper cache 3585 entries. 3586 3587 nfs.max_session_cb_slots= 3588 [NFSv4.1] Sets the maximum number of session 3589 slots the client will assign to the callback 3590 channel. This determines the maximum number of 3591 callbacks the client will process in parallel for 3592 a particular server. 3593 3594 nfs.max_session_slots= 3595 [NFSv4.1] Sets the maximum number of session slots 3596 the client will attempt to negotiate with the server. 3597 This limits the number of simultaneous RPC requests 3598 that the client can send to the NFSv4.1 server. 3599 Note that there is little point in setting this 3600 value higher than the max_tcp_slot_table_limit. 3601 3602 nfs.nfs4_disable_idmapping= 3603 [NFSv4] When set to the default of '1', this option 3604 ensures that both the RPC level authentication 3605 scheme and the NFS level operations agree to use 3606 numeric uids/gids if the mount is using the 3607 'sec=sys' security flavour. In effect it is 3608 disabling idmapping, which can make migration from 3609 legacy NFSv2/v3 systems to NFSv4 easier. 3610 Servers that do not support this mode of operation 3611 will be autodetected by the client, and it will fall 3612 back to using the idmapper. 3613 To turn off this behaviour, set the value to '0'. 3614 3615 nfs.nfs4_unique_id= 3616 [NFS4] Specify an additional fixed unique ident- 3617 ification string that NFSv4 clients can insert into 3618 their nfs_client_id4 string. This is typically a 3619 UUID that is generated at system install time. 3620 3621 nfs.recover_lost_locks= 3622 [NFSv4] Attempt to recover locks that were lost due 3623 to a lease timeout on the server. Please note that 3624 doing this risks data corruption, since there are 3625 no guarantees that the file will remain unchanged 3626 after the locks are lost. 3627 If you want to enable the kernel legacy behaviour of 3628 attempting to recover these locks, then set this 3629 parameter to '1'. 3630 The default parameter value of '0' causes the kernel 3631 not to attempt recovery of lost locks. 3632 3633 nfs.send_implementation_id= 3634 [NFSv4.1] Send client implementation identification 3635 information in exchange_id requests. 3636 If zero, no implementation identification information 3637 will be sent. 3638 The default is to send the implementation identification 3639 information. 3640 3641 nfs4.layoutstats_timer= 3642 [NFSv4.2] Change the rate at which the kernel sends 3643 layoutstats to the pNFS metadata server. 3644 3645 Setting this to value to 0 causes the kernel to use 3646 whatever value is the default set by the layout 3647 driver. A non-zero value sets the minimum interval 3648 in seconds between layoutstats transmissions. 3649 3650 nfsd.inter_copy_offload_enable= 3651 [NFSv4.2] When set to 1, the server will support 3652 server-to-server copies for which this server is 3653 the destination of the copy. 3654 3655 nfsd.nfs4_disable_idmapping= 3656 [NFSv4] When set to the default of '1', the NFSv4 3657 server will return only numeric uids and gids to 3658 clients using auth_sys, and will accept numeric uids 3659 and gids from such clients. This is intended to ease 3660 migration from NFSv2/v3. 3661 3662 nfsd.nfsd4_ssc_umount_timeout= 3663 [NFSv4.2] When used as the destination of a 3664 server-to-server copy, knfsd temporarily mounts 3665 the source server. It caches the mount in case 3666 it will be needed again, and discards it if not 3667 used for the number of milliseconds specified by 3668 this parameter. 3669 3670 nfsaddrs= [NFS] Deprecated. Use ip= instead. 3671 See Documentation/admin-guide/nfs/nfsroot.rst. 3672 3673 nfsroot= [NFS] nfs root filesystem for disk-less boxes. 3674 See Documentation/admin-guide/nfs/nfsroot.rst. 3675 3676 nfsrootdebug [NFS] enable nfsroot debugging messages. 3677 See Documentation/admin-guide/nfs/nfsroot.rst. 3678 3679 nmi_backtrace.backtrace_idle [KNL] 3680 Dump stacks even of idle CPUs in response to an 3681 NMI stack-backtrace request. 3682 3683 nmi_debug= [KNL,SH] Specify one or more actions to take 3684 when a NMI is triggered. 3685 Format: [state][,regs][,debounce][,die] 3686 3687 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels 3688 Format: [panic,][nopanic,][num] 3689 Valid num: 0 or 1 3690 0 - turn hardlockup detector in nmi_watchdog off 3691 1 - turn hardlockup detector in nmi_watchdog on 3692 When panic is specified, panic when an NMI watchdog 3693 timeout occurs (or 'nopanic' to not panic on an NMI 3694 watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set) 3695 To disable both hard and soft lockup detectors, 3696 please see 'nowatchdog'. 3697 This is useful when you use a panic=... timeout and 3698 need the box quickly up again. 3699 3700 These settings can be accessed at runtime via 3701 the nmi_watchdog and hardlockup_panic sysctls. 3702 3703 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths 3704 emulation library even if a 387 maths coprocessor 3705 is present. 3706 3707 no4lvl [RISCV] Disable 4-level and 5-level paging modes. Forces 3708 kernel to use 3-level paging instead. 3709 3710 no5lvl [X86-64,RISCV] Disable 5-level paging mode. Forces 3711 kernel to use 4-level paging instead. 3712 3713 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien 3714 caches in the slab allocator. Saves per-node memory, 3715 but will impact performance. 3716 3717 noalign [KNL,ARM] 3718 3719 noaltinstr [S390] Disables alternative instructions patching 3720 (CPU alternatives feature). 3721 3722 noapic [SMP,APIC] Tells the kernel to not make use of any 3723 IOAPICs that may be present in the system. 3724 3725 noautogroup Disable scheduler automatic task group creation. 3726 3727 nocache [ARM] 3728 3729 no_console_suspend 3730 [HW] Never suspend the console 3731 Disable suspending of consoles during suspend and 3732 hibernate operations. Once disabled, debugging 3733 messages can reach various consoles while the rest 3734 of the system is being put to sleep (ie, while 3735 debugging driver suspend/resume hooks). This may 3736 not work reliably with all consoles, but is known 3737 to work with serial and VGA consoles. 3738 To facilitate more flexible debugging, we also add 3739 console_suspend, a printk module parameter to control 3740 it. Users could use console_suspend (usually 3741 /sys/module/printk/parameters/console_suspend) to 3742 turn on/off it dynamically. 3743 3744 no_debug_objects 3745 [KNL] Disable object debugging 3746 3747 nodsp [SH] Disable hardware DSP at boot time. 3748 3749 noefi Disable EFI runtime services support. 3750 3751 no_entry_flush [PPC] Don't flush the L1-D cache when entering the kernel. 3752 3753 noexec [IA-64] 3754 3755 noexec32 [X86-64] 3756 This affects only 32-bit executables. 3757 noexec32=on: enable non-executable mappings (default) 3758 read doesn't imply executable mappings 3759 noexec32=off: disable non-executable mappings 3760 read implies executable mappings 3761 3762 no_file_caps Tells the kernel not to honor file capabilities. The 3763 only way then for a file to be executed with privilege 3764 is to be setuid root or executed by root. 3765 3766 nofpu [MIPS,SH] Disable hardware FPU at boot time. 3767 3768 nofsgsbase [X86] Disables FSGSBASE instructions. 3769 3770 nofxsr [BUGS=X86-32] Disables x86 floating point extended 3771 register save and restore. The kernel will only save 3772 legacy floating-point registers on task switch. 3773 3774 nohalt [IA-64] Tells the kernel not to use the power saving 3775 function PAL_HALT_LIGHT when idle. This increases 3776 power-consumption. On the positive side, it reduces 3777 interrupt wake-up latency, which may improve performance 3778 in certain environments such as networked servers or 3779 real-time systems. 3780 3781 no_hash_pointers 3782 Force pointers printed to the console or buffers to be 3783 unhashed. By default, when a pointer is printed via %p 3784 format string, that pointer is "hashed", i.e. obscured 3785 by hashing the pointer value. This is a security feature 3786 that hides actual kernel addresses from unprivileged 3787 users, but it also makes debugging the kernel more 3788 difficult since unequal pointers can no longer be 3789 compared. However, if this command-line option is 3790 specified, then all normal pointers will have their true 3791 value printed. This option should only be specified when 3792 debugging the kernel. Please do not use on production 3793 kernels. 3794 3795 nohibernate [HIBERNATION] Disable hibernation and resume. 3796 3797 nohlt [ARM,ARM64,MICROBLAZE,MIPS,PPC,SH] Forces the kernel to 3798 busy wait in do_idle() and not use the arch_cpu_idle() 3799 implementation; requires CONFIG_GENERIC_IDLE_POLL_SETUP 3800 to be effective. This is useful on platforms where the 3801 sleep(SH) or wfi(ARM,ARM64) instructions do not work 3802 correctly or when doing power measurements to evaluate 3803 the impact of the sleep instructions. This is also 3804 useful when using JTAG debugger. 3805 3806 nohugeiomap [KNL,X86,PPC,ARM64] Disable kernel huge I/O mappings. 3807 3808 nohugevmalloc [KNL,X86,PPC,ARM64] Disable kernel huge vmalloc mappings. 3809 3810 nohz= [KNL] Boottime enable/disable dynamic ticks 3811 Valid arguments: on, off 3812 Default: on 3813 3814 nohz_full= [KNL,BOOT,SMP,ISOL] 3815 The argument is a cpu list, as described above. 3816 In kernels built with CONFIG_NO_HZ_FULL=y, set 3817 the specified list of CPUs whose tick will be stopped 3818 whenever possible. The boot CPU will be forced outside 3819 the range to maintain the timekeeping. Any CPUs 3820 in this list will have their RCU callbacks offloaded, 3821 just as if they had also been called out in the 3822 rcu_nocbs= boot parameter. 3823 3824 Note that this argument takes precedence over 3825 the CONFIG_RCU_NOCB_CPU_DEFAULT_ALL option. 3826 3827 noinitrd [RAM] Tells the kernel not to load any configured 3828 initial RAM disk. 3829 3830 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt 3831 remapping. 3832 [Deprecated - use intremap=off] 3833 3834 nointroute [IA-64] 3835 3836 noinvpcid [X86] Disable the INVPCID cpu feature. 3837 3838 noiotrap [SH] Disables trapped I/O port accesses. 3839 3840 noirqdebug [X86-32] Disables the code which attempts to detect and 3841 disable unhandled interrupt sources. 3842 3843 noisapnp [ISAPNP] Disables ISA PnP code. 3844 3845 nojitter [IA-64] Disables jitter checking for ITC timers. 3846 3847 nokaslr [KNL] 3848 When CONFIG_RANDOMIZE_BASE is set, this disables 3849 kernel and module base offset ASLR (Address Space 3850 Layout Randomization). 3851 3852 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page 3853 fault handling. 3854 3855 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver 3856 3857 nolapic [X86-32,APIC] Do not enable or use the local APIC. 3858 3859 nolapic_timer [X86-32,APIC] Do not use the local APIC timer. 3860 3861 nomca [IA-64] Disable machine check abort handling 3862 3863 nomce [X86-32] Disable Machine Check Exception 3864 3865 nomfgpt [X86-32] Disable Multi-Function General Purpose 3866 Timer usage (for AMD Geode machines). 3867 3868 nomodeset Disable kernel modesetting. Most systems' firmware 3869 sets up a display mode and provides framebuffer memory 3870 for output. With nomodeset, DRM and fbdev drivers will 3871 not load if they could possibly displace the pre- 3872 initialized output. Only the system framebuffer will 3873 be available for use. The respective drivers will not 3874 perform display-mode changes or accelerated rendering. 3875 3876 Useful as error fallback, or for testing and debugging. 3877 3878 nomodule Disable module load 3879 3880 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to 3881 shutdown the other cpus. Instead use the REBOOT_VECTOR 3882 irq. 3883 3884 nopat [X86] Disable PAT (page attribute table extension of 3885 pagetables) support. 3886 3887 nopcid [X86-64] Disable the PCID cpu feature. 3888 3889 nopku [X86] Disable Memory Protection Keys CPU feature found 3890 in some Intel CPUs. 3891 3892 nopti [X86-64] 3893 Equivalent to pti=off 3894 3895 nopv= [X86,XEN,KVM,HYPER_V,VMWARE] 3896 Disables the PV optimizations forcing the guest to run 3897 as generic guest with no PV drivers. Currently support 3898 XEN HVM, KVM, HYPER_V and VMWARE guest. 3899 3900 nopvspin [X86,XEN,KVM] 3901 Disables the qspinlock slow path using PV optimizations 3902 which allow the hypervisor to 'idle' the guest on lock 3903 contention. 3904 3905 norandmaps Don't use address space randomization. Equivalent to 3906 echo 0 > /proc/sys/kernel/randomize_va_space 3907 3908 noreplace-smp [X86-32,SMP] Don't replace SMP instructions 3909 with UP alternatives 3910 3911 noresume [SWSUSP] Disables resume and restores original swap 3912 space. 3913 3914 nosbagart [IA-64] 3915 3916 no-scroll [VGA] Disables scrollback. 3917 This is required for the Braillex ib80-piezo Braille 3918 reader made by F.H. Papenmeier (Germany). 3919 3920 nosgx [X86-64,SGX] Disables Intel SGX kernel support. 3921 3922 nosmap [PPC] 3923 Disable SMAP (Supervisor Mode Access Prevention) 3924 even if it is supported by processor. 3925 3926 nosmep [PPC64s] 3927 Disable SMEP (Supervisor Mode Execution Prevention) 3928 even if it is supported by processor. 3929 3930 nosmp [SMP] Tells an SMP kernel to act as a UP kernel, 3931 and disable the IO APIC. legacy for "maxcpus=0". 3932 3933 nosmt [KNL,MIPS,PPC,S390] Disable symmetric multithreading (SMT). 3934 Equivalent to smt=1. 3935 3936 [KNL,X86,PPC] Disable symmetric multithreading (SMT). 3937 nosmt=force: Force disable SMT, cannot be undone 3938 via the sysfs control file. 3939 3940 nosoftlockup [KNL] Disable the soft-lockup detector. 3941 3942 nospec_store_bypass_disable 3943 [HW] Disable all mitigations for the Speculative Store Bypass vulnerability 3944 3945 nospectre_bhb [ARM64] Disable all mitigations for Spectre-BHB (branch 3946 history injection) vulnerability. System may allow data leaks 3947 with this option. 3948 3949 nospectre_v1 [X86,PPC] Disable mitigations for Spectre Variant 1 3950 (bounds check bypass). With this option data leaks are 3951 possible in the system. 3952 3953 nospectre_v2 [X86,PPC_E500,ARM64] Disable all mitigations for 3954 the Spectre variant 2 (indirect branch prediction) 3955 vulnerability. System may allow data leaks with this 3956 option. 3957 3958 no-steal-acc [X86,PV_OPS,ARM64,PPC/PSERIES] Disable paravirtualized 3959 steal time accounting. steal time is computed, but 3960 won't influence scheduler behaviour 3961 3962 nosync [HW,M68K] Disables sync negotiation for all devices. 3963 3964 no_timer_check [X86,APIC] Disables the code which tests for 3965 broken timer IRQ sources. 3966 3967 no_uaccess_flush 3968 [PPC] Don't flush the L1-D cache after accessing user data. 3969 3970 novmcoredd [KNL,KDUMP] 3971 Disable device dump. Device dump allows drivers to 3972 append dump data to vmcore so you can collect driver 3973 specified debug info. Drivers can append the data 3974 without any limit and this data is stored in memory, 3975 so this may cause significant memory stress. Disabling 3976 device dump can help save memory but the driver debug 3977 data will be no longer available. This parameter 3978 is only available when CONFIG_PROC_VMCORE_DEVICE_DUMP 3979 is set. 3980 3981 no-vmw-sched-clock 3982 [X86,PV_OPS] Disable paravirtualized VMware scheduler 3983 clock and use the default one. 3984 3985 nowatchdog [KNL] Disable both lockup detectors, i.e. 3986 soft-lockup and NMI watchdog (hard-lockup). 3987 3988 nowb [ARM] 3989 3990 nox2apic [X86-64,APIC] Do not enable x2APIC mode. 3991 3992 NOTE: this parameter will be ignored on systems with the 3993 LEGACY_XAPIC_DISABLED bit set in the 3994 IA32_XAPIC_DISABLE_STATUS MSR. 3995 3996 noxsave [BUGS=X86] Disables x86 extended register state save 3997 and restore using xsave. The kernel will fallback to 3998 enabling legacy floating-point and sse state. 3999 4000 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended 4001 register states. The kernel will fall back to use 4002 xsave to save the states. By using this parameter, 4003 performance of saving the states is degraded because 4004 xsave doesn't support modified optimization while 4005 xsaveopt supports it on xsaveopt enabled systems. 4006 4007 noxsaves [X86] Disables xsaves and xrstors used in saving and 4008 restoring x86 extended register state in compacted 4009 form of xsave area. The kernel will fall back to use 4010 xsaveopt and xrstor to save and restore the states 4011 in standard form of xsave area. By using this 4012 parameter, xsave area per process might occupy more 4013 memory on xsaves enabled systems. 4014 4015 nps_mtm_hs_ctr= [KNL,ARC] 4016 This parameter sets the maximum duration, in 4017 cycles, each HW thread of the CTOP can run 4018 without interruptions, before HW switches it. 4019 The actual maximum duration is 16 times this 4020 parameter's value. 4021 Format: integer between 1 and 255 4022 Default: 255 4023 4024 nptcg= [IA-64] Override max number of concurrent global TLB 4025 purges which is reported from either PAL_VM_SUMMARY or 4026 SAL PALO. 4027 4028 nr_cpus= [SMP] Maximum number of processors that an SMP kernel 4029 could support. nr_cpus=n : n >= 1 limits the kernel to 4030 support 'n' processors. It could be larger than the 4031 number of already plugged CPU during bootup, later in 4032 runtime you can physically add extra cpu until it reaches 4033 n. So during boot up some boot time memory for per-cpu 4034 variables need be pre-allocated for later physical cpu 4035 hot plugging. 4036 4037 nr_uarts= [SERIAL] maximum number of UARTs to be registered. 4038 4039 numa=off [KNL, ARM64, PPC, RISCV, SPARC, X86] Disable NUMA, Only 4040 set up a single NUMA node spanning all memory. 4041 4042 numa_balancing= [KNL,ARM64,PPC,RISCV,S390,X86] Enable or disable automatic 4043 NUMA balancing. 4044 Allowed values are enable and disable 4045 4046 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA. 4047 'node', 'default' can be specified 4048 This can be set from sysctl after boot. 4049 See Documentation/admin-guide/sysctl/vm.rst for details. 4050 4051 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver. 4052 See Documentation/core-api/debugging-via-ohci1394.rst for more 4053 info. 4054 4055 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands 4056 Rather than timing out after 20 ms if an EC 4057 command is not properly ACKed, override the length 4058 of the timeout. We have interrupts disabled while 4059 waiting for the ACK, so if this is set too high 4060 interrupts *may* be lost! 4061 4062 omap_mux= [OMAP] Override bootloader pin multiplexing. 4063 Format: <mux_mode0.mode_name=value>... 4064 For example, to override I2C bus2: 4065 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100 4066 4067 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration 4068 4069 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock] 4070 4071 boundary - index of last SLC block on Flex-OneNAND. 4072 The remaining blocks are configured as MLC blocks. 4073 lock - Configure if Flex-OneNAND boundary should be locked. 4074 Once locked, the boundary cannot be changed. 4075 1 indicates lock status, 0 indicates unlock status. 4076 4077 oops=panic Always panic on oopses. Default is to just kill the 4078 process, but there is a small probability of 4079 deadlocking the machine. 4080 This will also cause panics on machine check exceptions. 4081 Useful together with panic=30 to trigger a reboot. 4082 4083 page_alloc.shuffle= 4084 [KNL] Boolean flag to control whether the page allocator 4085 should randomize its free lists. The randomization may 4086 be automatically enabled if the kernel detects it is 4087 running on a platform with a direct-mapped memory-side 4088 cache, and this parameter can be used to 4089 override/disable that behavior. The state of the flag 4090 can be read from sysfs at: 4091 /sys/module/page_alloc/parameters/shuffle. 4092 4093 page_owner= [KNL] Boot-time page_owner enabling option. 4094 Storage of the information about who allocated 4095 each page is disabled in default. With this switch, 4096 we can turn it on. 4097 on: enable the feature 4098 4099 page_poison= [KNL] Boot-time parameter changing the state of 4100 poisoning on the buddy allocator, available with 4101 CONFIG_PAGE_POISONING=y. 4102 off: turn off poisoning (default) 4103 on: turn on poisoning 4104 4105 page_reporting.page_reporting_order= 4106 [KNL] Minimal page reporting order 4107 Format: <integer> 4108 Adjust the minimal page reporting order. The page 4109 reporting is disabled when it exceeds MAX_ORDER. 4110 4111 panic= [KNL] Kernel behaviour on panic: delay <timeout> 4112 timeout > 0: seconds before rebooting 4113 timeout = 0: wait forever 4114 timeout < 0: reboot immediately 4115 Format: <timeout> 4116 4117 panic_on_taint= Bitmask for conditionally calling panic() in add_taint() 4118 Format: <hex>[,nousertaint] 4119 Hexadecimal bitmask representing the set of TAINT flags 4120 that will cause the kernel to panic when add_taint() is 4121 called with any of the flags in this set. 4122 The optional switch "nousertaint" can be utilized to 4123 prevent userspace forced crashes by writing to sysctl 4124 /proc/sys/kernel/tainted any flagset matching with the 4125 bitmask set on panic_on_taint. 4126 See Documentation/admin-guide/tainted-kernels.rst for 4127 extra details on the taint flags that users can pick 4128 to compose the bitmask to assign to panic_on_taint. 4129 4130 panic_on_warn=1 panic() instead of WARN(). Useful to cause kdump 4131 on a WARN(). 4132 4133 panic_print= Bitmask for printing system info when panic happens. 4134 User can chose combination of the following bits: 4135 bit 0: print all tasks info 4136 bit 1: print system memory info 4137 bit 2: print timer info 4138 bit 3: print locks info if CONFIG_LOCKDEP is on 4139 bit 4: print ftrace buffer 4140 bit 5: print all printk messages in buffer 4141 bit 6: print all CPUs backtrace (if available in the arch) 4142 *Be aware* that this option may print a _lot_ of lines, 4143 so there are risks of losing older messages in the log. 4144 Use this option carefully, maybe worth to setup a 4145 bigger log buffer with "log_buf_len" along with this. 4146 4147 parkbd.port= [HW] Parallel port number the keyboard adapter is 4148 connected to, default is 0. 4149 Format: <parport#> 4150 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation, 4151 0 for XT, 1 for AT (default is AT). 4152 Format: <mode> 4153 4154 parport= [HW,PPT] Specify parallel ports. 0 disables. 4155 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] } 4156 Use 'auto' to force the driver to use any 4157 IRQ/DMA settings detected (the default is to 4158 ignore detected IRQ/DMA settings because of 4159 possible conflicts). You can specify the base 4160 address, IRQ, and DMA settings; IRQ and DMA 4161 should be numbers, or 'auto' (for using detected 4162 settings on that particular port), or 'nofifo' 4163 (to avoid using a FIFO even if it is detected). 4164 Parallel ports are assigned in the order they 4165 are specified on the command line, starting 4166 with parport0. 4167 4168 parport_init_mode= [HW,PPT] 4169 Configure VIA parallel port to operate in 4170 a specific mode. This is necessary on Pegasos 4171 computer where firmware has no options for setting 4172 up parallel port mode and sets it to spp. 4173 Currently this function knows 686a and 8231 chips. 4174 Format: [spp|ps2|epp|ecp|ecpepp] 4175 4176 pata_legacy.all= [HW,LIBATA] 4177 Format: <int> 4178 Set to non-zero to probe primary and secondary ISA 4179 port ranges on PCI systems where no PCI PATA device 4180 has been found at either range. Disabled by default. 4181 4182 pata_legacy.autospeed= [HW,LIBATA] 4183 Format: <int> 4184 Set to non-zero if a chip is present that snoops speed 4185 changes. Disabled by default. 4186 4187 pata_legacy.ht6560a= [HW,LIBATA] 4188 Format: <int> 4189 Set to 1, 2, or 3 for HT 6560A on the primary channel, 4190 the secondary channel, or both channels respectively. 4191 Disabled by default. 4192 4193 pata_legacy.ht6560b= [HW,LIBATA] 4194 Format: <int> 4195 Set to 1, 2, or 3 for HT 6560B on the primary channel, 4196 the secondary channel, or both channels respectively. 4197 Disabled by default. 4198 4199 pata_legacy.iordy_mask= [HW,LIBATA] 4200 Format: <int> 4201 IORDY enable mask. Set individual bits to allow IORDY 4202 for the respective channel. Bit 0 is for the first 4203 legacy channel handled by this driver, bit 1 is for 4204 the second channel, and so on. The sequence will often 4205 correspond to the primary legacy channel, the secondary 4206 legacy channel, and so on, but the handling of a PCI 4207 bus and the use of other driver options may interfere 4208 with the sequence. By default IORDY is allowed across 4209 all channels. 4210 4211 pata_legacy.opti82c46x= [HW,LIBATA] 4212 Format: <int> 4213 Set to 1, 2, or 3 for Opti 82c611A on the primary 4214 channel, the secondary channel, or both channels 4215 respectively. Disabled by default. 4216 4217 pata_legacy.opti82c611a= [HW,LIBATA] 4218 Format: <int> 4219 Set to 1, 2, or 3 for Opti 82c465MV on the primary 4220 channel, the secondary channel, or both channels 4221 respectively. Disabled by default. 4222 4223 pata_legacy.pio_mask= [HW,LIBATA] 4224 Format: <int> 4225 PIO mode mask for autospeed devices. Set individual 4226 bits to allow the use of the respective PIO modes. 4227 Bit 0 is for mode 0, bit 1 is for mode 1, and so on. 4228 All modes allowed by default. 4229 4230 pata_legacy.probe_all= [HW,LIBATA] 4231 Format: <int> 4232 Set to non-zero to probe tertiary and further ISA 4233 port ranges on PCI systems. Disabled by default. 4234 4235 pata_legacy.probe_mask= [HW,LIBATA] 4236 Format: <int> 4237 Probe mask for legacy ISA PATA ports. Depending on 4238 platform configuration and the use of other driver 4239 options up to 6 legacy ports are supported: 0x1f0, 4240 0x170, 0x1e8, 0x168, 0x1e0, 0x160, however probing 4241 of individual ports can be disabled by setting the 4242 corresponding bits in the mask to 1. Bit 0 is for 4243 the first port in the list above (0x1f0), and so on. 4244 By default all supported ports are probed. 4245 4246 pata_legacy.qdi= [HW,LIBATA] 4247 Format: <int> 4248 Set to non-zero to probe QDI controllers. By default 4249 set to 1 if CONFIG_PATA_QDI_MODULE, 0 otherwise. 4250 4251 pata_legacy.winbond= [HW,LIBATA] 4252 Format: <int> 4253 Set to non-zero to probe Winbond controllers. Use 4254 the standard I/O port (0x130) if 1, otherwise the 4255 value given is the I/O port to use (typically 0x1b0). 4256 By default set to 1 if CONFIG_PATA_WINBOND_VLB_MODULE, 4257 0 otherwise. 4258 4259 pata_platform.pio_mask= [HW,LIBATA] 4260 Format: <int> 4261 Supported PIO mode mask. Set individual bits to allow 4262 the use of the respective PIO modes. Bit 0 is for 4263 mode 0, bit 1 is for mode 1, and so on. Mode 0 only 4264 allowed by default. 4265 4266 pause_on_oops=<int> 4267 Halt all CPUs after the first oops has been printed for 4268 the specified number of seconds. This is to be used if 4269 your oopses keep scrolling off the screen. 4270 4271 pcbit= [HW,ISDN] 4272 4273 pci=option[,option...] [PCI] various PCI subsystem options. 4274 4275 Some options herein operate on a specific device 4276 or a set of devices (<pci_dev>). These are 4277 specified in one of the following formats: 4278 4279 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]* 4280 pci:<vendor>:<device>[:<subvendor>:<subdevice>] 4281 4282 Note: the first format specifies a PCI 4283 bus/device/function address which may change 4284 if new hardware is inserted, if motherboard 4285 firmware changes, or due to changes caused 4286 by other kernel parameters. If the 4287 domain is left unspecified, it is 4288 taken to be zero. Optionally, a path 4289 to a device through multiple device/function 4290 addresses can be specified after the base 4291 address (this is more robust against 4292 renumbering issues). The second format 4293 selects devices using IDs from the 4294 configuration space which may match multiple 4295 devices in the system. 4296 4297 earlydump dump PCI config space before the kernel 4298 changes anything 4299 off [X86] don't probe for the PCI bus 4300 bios [X86-32] force use of PCI BIOS, don't access 4301 the hardware directly. Use this if your machine 4302 has a non-standard PCI host bridge. 4303 nobios [X86-32] disallow use of PCI BIOS, only direct 4304 hardware access methods are allowed. Use this 4305 if you experience crashes upon bootup and you 4306 suspect they are caused by the BIOS. 4307 conf1 [X86] Force use of PCI Configuration Access 4308 Mechanism 1 (config address in IO port 0xCF8, 4309 data in IO port 0xCFC, both 32-bit). 4310 conf2 [X86] Force use of PCI Configuration Access 4311 Mechanism 2 (IO port 0xCF8 is an 8-bit port for 4312 the function, IO port 0xCFA, also 8-bit, sets 4313 bus number. The config space is then accessed 4314 through ports 0xC000-0xCFFF). 4315 See http://wiki.osdev.org/PCI for more info 4316 on the configuration access mechanisms. 4317 noaer [PCIE] If the PCIEAER kernel config parameter is 4318 enabled, this kernel boot option can be used to 4319 disable the use of PCIE advanced error reporting. 4320 nodomains [PCI] Disable support for multiple PCI 4321 root domains (aka PCI segments, in ACPI-speak). 4322 nommconf [X86] Disable use of MMCONFIG for PCI 4323 Configuration 4324 check_enable_amd_mmconf [X86] check for and enable 4325 properly configured MMIO access to PCI 4326 config space on AMD family 10h CPU 4327 nomsi [MSI] If the PCI_MSI kernel config parameter is 4328 enabled, this kernel boot option can be used to 4329 disable the use of MSI interrupts system-wide. 4330 noioapicquirk [APIC] Disable all boot interrupt quirks. 4331 Safety option to keep boot IRQs enabled. This 4332 should never be necessary. 4333 ioapicreroute [APIC] Enable rerouting of boot IRQs to the 4334 primary IO-APIC for bridges that cannot disable 4335 boot IRQs. This fixes a source of spurious IRQs 4336 when the system masks IRQs. 4337 noioapicreroute [APIC] Disable workaround that uses the 4338 boot IRQ equivalent of an IRQ that connects to 4339 a chipset where boot IRQs cannot be disabled. 4340 The opposite of ioapicreroute. 4341 biosirq [X86-32] Use PCI BIOS calls to get the interrupt 4342 routing table. These calls are known to be buggy 4343 on several machines and they hang the machine 4344 when used, but on other computers it's the only 4345 way to get the interrupt routing table. Try 4346 this option if the kernel is unable to allocate 4347 IRQs or discover secondary PCI buses on your 4348 motherboard. 4349 rom [X86] Assign address space to expansion ROMs. 4350 Use with caution as certain devices share 4351 address decoders between ROMs and other 4352 resources. 4353 norom [X86] Do not assign address space to 4354 expansion ROMs that do not already have 4355 BIOS assigned address ranges. 4356 nobar [X86] Do not assign address space to the 4357 BARs that weren't assigned by the BIOS. 4358 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be 4359 assigned automatically to PCI devices. You can 4360 make the kernel exclude IRQs of your ISA cards 4361 this way. 4362 pirqaddr=0xAAAAA [X86] Specify the physical address 4363 of the PIRQ table (normally generated 4364 by the BIOS) if it is outside the 4365 F0000h-100000h range. 4366 lastbus=N [X86] Scan all buses thru bus #N. Can be 4367 useful if the kernel is unable to find your 4368 secondary buses and you want to tell it 4369 explicitly which ones they are. 4370 assign-busses [X86] Always assign all PCI bus 4371 numbers ourselves, overriding 4372 whatever the firmware may have done. 4373 usepirqmask [X86] Honor the possible IRQ mask stored 4374 in the BIOS $PIR table. This is needed on 4375 some systems with broken BIOSes, notably 4376 some HP Pavilion N5400 and Omnibook XE3 4377 notebooks. This will have no effect if ACPI 4378 IRQ routing is enabled. 4379 noacpi [X86] Do not use ACPI for IRQ routing 4380 or for PCI scanning. 4381 use_crs [X86] Use PCI host bridge window information 4382 from ACPI. On BIOSes from 2008 or later, this 4383 is enabled by default. If you need to use this, 4384 please report a bug. 4385 nocrs [X86] Ignore PCI host bridge windows from ACPI. 4386 If you need to use this, please report a bug. 4387 use_e820 [X86] Use E820 reservations to exclude parts of 4388 PCI host bridge windows. This is a workaround 4389 for BIOS defects in host bridge _CRS methods. 4390 If you need to use this, please report a bug to 4391 <linux-pci@vger.kernel.org>. 4392 no_e820 [X86] Ignore E820 reservations for PCI host 4393 bridge windows. This is the default on modern 4394 hardware. If you need to use this, please report 4395 a bug to <linux-pci@vger.kernel.org>. 4396 routeirq Do IRQ routing for all PCI devices. 4397 This is normally done in pci_enable_device(), 4398 so this option is a temporary workaround 4399 for broken drivers that don't call it. 4400 skip_isa_align [X86] do not align io start addr, so can 4401 handle more pci cards 4402 noearly [X86] Don't do any early type 1 scanning. 4403 This might help on some broken boards which 4404 machine check when some devices' config space 4405 is read. But various workarounds are disabled 4406 and some IOMMU drivers will not work. 4407 bfsort Sort PCI devices into breadth-first order. 4408 This sorting is done to get a device 4409 order compatible with older (<= 2.4) kernels. 4410 nobfsort Don't sort PCI devices into breadth-first order. 4411 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size) 4412 tuning and use the BIOS-configured MPS defaults. 4413 pcie_bus_safe Set every device's MPS to the largest value 4414 supported by all devices below the root complex. 4415 pcie_bus_perf Set device MPS to the largest allowable MPS 4416 based on its parent bus. Also set MRRS (Max 4417 Read Request Size) to the largest supported 4418 value (no larger than the MPS that the device 4419 or bus can support) for best performance. 4420 pcie_bus_peer2peer Set every device's MPS to 128B, which 4421 every device is guaranteed to support. This 4422 configuration allows peer-to-peer DMA between 4423 any pair of devices, possibly at the cost of 4424 reduced performance. This also guarantees 4425 that hot-added devices will work. 4426 cbiosize=nn[KMG] The fixed amount of bus space which is 4427 reserved for the CardBus bridge's IO window. 4428 The default value is 256 bytes. 4429 cbmemsize=nn[KMG] The fixed amount of bus space which is 4430 reserved for the CardBus bridge's memory 4431 window. The default value is 64 megabytes. 4432 resource_alignment= 4433 Format: 4434 [<order of align>@]<pci_dev>[; ...] 4435 Specifies alignment and device to reassign 4436 aligned memory resources. How to 4437 specify the device is described above. 4438 If <order of align> is not specified, 4439 PAGE_SIZE is used as alignment. 4440 A PCI-PCI bridge can be specified if resource 4441 windows need to be expanded. 4442 To specify the alignment for several 4443 instances of a device, the PCI vendor, 4444 device, subvendor, and subdevice may be 4445 specified, e.g., 12@pci:8086:9c22:103c:198f 4446 for 4096-byte alignment. 4447 ecrc= Enable/disable PCIe ECRC (transaction layer 4448 end-to-end CRC checking). Only effective if 4449 OS has native AER control (either granted by 4450 ACPI _OSC or forced via "pcie_ports=native") 4451 bios: Use BIOS/firmware settings. This is the 4452 the default. 4453 off: Turn ECRC off 4454 on: Turn ECRC on. 4455 hpiosize=nn[KMG] The fixed amount of bus space which is 4456 reserved for hotplug bridge's IO window. 4457 Default size is 256 bytes. 4458 hpmmiosize=nn[KMG] The fixed amount of bus space which is 4459 reserved for hotplug bridge's MMIO window. 4460 Default size is 2 megabytes. 4461 hpmmioprefsize=nn[KMG] The fixed amount of bus space which is 4462 reserved for hotplug bridge's MMIO_PREF window. 4463 Default size is 2 megabytes. 4464 hpmemsize=nn[KMG] The fixed amount of bus space which is 4465 reserved for hotplug bridge's MMIO and 4466 MMIO_PREF window. 4467 Default size is 2 megabytes. 4468 hpbussize=nn The minimum amount of additional bus numbers 4469 reserved for buses below a hotplug bridge. 4470 Default is 1. 4471 realloc= Enable/disable reallocating PCI bridge resources 4472 if allocations done by BIOS are too small to 4473 accommodate resources required by all child 4474 devices. 4475 off: Turn realloc off 4476 on: Turn realloc on 4477 realloc same as realloc=on 4478 noari do not use PCIe ARI. 4479 noats [PCIE, Intel-IOMMU, AMD-IOMMU] 4480 do not use PCIe ATS (and IOMMU device IOTLB). 4481 pcie_scan_all Scan all possible PCIe devices. Otherwise we 4482 only look for one device below a PCIe downstream 4483 port. 4484 big_root_window Try to add a big 64bit memory window to the PCIe 4485 root complex on AMD CPUs. Some GFX hardware 4486 can resize a BAR to allow access to all VRAM. 4487 Adding the window is slightly risky (it may 4488 conflict with unreported devices), so this 4489 taints the kernel. 4490 disable_acs_redir=<pci_dev>[; ...] 4491 Specify one or more PCI devices (in the format 4492 specified above) separated by semicolons. 4493 Each device specified will have the PCI ACS 4494 redirect capabilities forced off which will 4495 allow P2P traffic between devices through 4496 bridges without forcing it upstream. Note: 4497 this removes isolation between devices and 4498 may put more devices in an IOMMU group. 4499 force_floating [S390] Force usage of floating interrupts. 4500 nomio [S390] Do not use MIO instructions. 4501 norid [S390] ignore the RID field and force use of 4502 one PCI domain per PCI function 4503 4504 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power 4505 Management. 4506 off Disable ASPM. 4507 force Enable ASPM even on devices that claim not to support it. 4508 WARNING: Forcing ASPM on may cause system lockups. 4509 4510 pcie_ports= [PCIE] PCIe port services handling: 4511 native Use native PCIe services (PME, AER, DPC, PCIe hotplug) 4512 even if the platform doesn't give the OS permission to 4513 use them. This may cause conflicts if the platform 4514 also tries to use these services. 4515 dpc-native Use native PCIe service for DPC only. May 4516 cause conflicts if firmware uses AER or DPC. 4517 compat Disable native PCIe services (PME, AER, DPC, PCIe 4518 hotplug). 4519 4520 pcie_port_pm= [PCIE] PCIe port power management handling: 4521 off Disable power management of all PCIe ports 4522 force Forcibly enable power management of all PCIe ports 4523 4524 pcie_pme= [PCIE,PM] Native PCIe PME signaling options: 4525 nomsi Do not use MSI for native PCIe PME signaling (this makes 4526 all PCIe root ports use INTx for all services). 4527 4528 pcmv= [HW,PCMCIA] BadgePAD 4 4529 4530 pd_ignore_unused 4531 [PM] 4532 Keep all power-domains already enabled by bootloader on, 4533 even if no driver has claimed them. This is useful 4534 for debug and development, but should not be 4535 needed on a platform with proper driver support. 4536 4537 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at 4538 boot time. 4539 Format: { 0 | 1 } 4540 See arch/parisc/kernel/pdc_chassis.c 4541 4542 percpu_alloc= Select which percpu first chunk allocator to use. 4543 Currently supported values are "embed" and "page". 4544 Archs may support subset or none of the selections. 4545 See comments in mm/percpu.c for details on each 4546 allocator. This parameter is primarily for debugging 4547 and performance comparison. 4548 4549 pirq= [SMP,APIC] Manual mp-table setup 4550 See Documentation/arch/x86/i386/IO-APIC.rst. 4551 4552 plip= [PPT,NET] Parallel port network link 4553 Format: { parport<nr> | timid | 0 } 4554 See also Documentation/admin-guide/parport.rst. 4555 4556 pmtmr= [X86] Manual setup of pmtmr I/O Port. 4557 Override pmtimer IOPort with a hex value. 4558 e.g. pmtmr=0x508 4559 4560 pmu_override= [PPC] Override the PMU. 4561 This option takes over the PMU facility, so it is no 4562 longer usable by perf. Setting this option starts the 4563 PMU counters by setting MMCR0 to 0 (the FC bit is 4564 cleared). If a number is given, then MMCR1 is set to 4565 that number, otherwise (e.g., 'pmu_override=on'), MMCR1 4566 remains 0. 4567 4568 pm_debug_messages [SUSPEND,KNL] 4569 Enable suspend/resume debug messages during boot up. 4570 4571 pnp.debug=1 [PNP] 4572 Enable PNP debug messages (depends on the 4573 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time 4574 via /sys/module/pnp/parameters/debug. We always show 4575 current resource usage; turning this on also shows 4576 possible settings and some assignment information. 4577 4578 pnpacpi= [ACPI] 4579 { off } 4580 4581 pnpbios= [ISAPNP] 4582 { on | off | curr | res | no-curr | no-res } 4583 4584 pnp_reserve_irq= 4585 [ISAPNP] Exclude IRQs for the autoconfiguration 4586 4587 pnp_reserve_dma= 4588 [ISAPNP] Exclude DMAs for the autoconfiguration 4589 4590 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration 4591 Ranges are in pairs (I/O port base and size). 4592 4593 pnp_reserve_mem= 4594 [ISAPNP] Exclude memory regions for the 4595 autoconfiguration. 4596 Ranges are in pairs (memory base and size). 4597 4598 ports= [IP_VS_FTP] IPVS ftp helper module 4599 Default is 21. 4600 Up to 8 (IP_VS_APP_MAX_PORTS) ports 4601 may be specified. 4602 Format: <port>,<port>.... 4603 4604 powersave=off [PPC] This option disables power saving features. 4605 It specifically disables cpuidle and sets the 4606 platform machine description specific power_save 4607 function to NULL. On Idle the CPU just reduces 4608 execution priority. 4609 4610 ppc_strict_facility_enable 4611 [PPC] This option catches any kernel floating point, 4612 Altivec, VSX and SPE outside of regions specifically 4613 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()). 4614 There is some performance impact when enabling this. 4615 4616 ppc_tm= [PPC] 4617 Format: {"off"} 4618 Disable Hardware Transactional Memory 4619 4620 preempt= [KNL] 4621 Select preemption mode if you have CONFIG_PREEMPT_DYNAMIC 4622 none - Limited to cond_resched() calls 4623 voluntary - Limited to cond_resched() and might_sleep() calls 4624 full - Any section that isn't explicitly preempt disabled 4625 can be preempted anytime. 4626 4627 print-fatal-signals= 4628 [KNL] debug: print fatal signals 4629 4630 If enabled, warn about various signal handling 4631 related application anomalies: too many signals, 4632 too many POSIX.1 timers, fatal signals causing a 4633 coredump - etc. 4634 4635 If you hit the warning due to signal overflow, 4636 you might want to try "ulimit -i unlimited". 4637 4638 default: off. 4639 4640 printk.always_kmsg_dump= 4641 Trigger kmsg_dump for cases other than kernel oops or 4642 panics 4643 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 4644 default: disabled 4645 4646 printk.console_no_auto_verbose= 4647 Disable console loglevel raise on oops, panic 4648 or lockdep-detected issues (only if lock debug is on). 4649 With an exception to setups with low baudrate on 4650 serial console, keeping this 0 is a good choice 4651 in order to provide more debug information. 4652 Format: <bool> 4653 default: 0 (auto_verbose is enabled) 4654 4655 printk.devkmsg={on,off,ratelimit} 4656 Control writing to /dev/kmsg. 4657 on - unlimited logging to /dev/kmsg from userspace 4658 off - logging to /dev/kmsg disabled 4659 ratelimit - ratelimit the logging 4660 Default: ratelimit 4661 4662 printk.time= Show timing data prefixed to each printk message line 4663 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 4664 4665 proc_mem.force_override= [KNL] 4666 Format: {always | ptrace | never} 4667 Traditionally /proc/pid/mem allows memory permissions to be 4668 overridden without restrictions. This option may be set to 4669 restrict that. Can be one of: 4670 - 'always': traditional behavior always allows mem overrides. 4671 - 'ptrace': only allow mem overrides for active ptracers. 4672 - 'never': never allow mem overrides. 4673 If not specified, default is the CONFIG_PROC_MEM_* choice. 4674 4675 processor.max_cstate= [HW,ACPI] 4676 Limit processor to maximum C-state 4677 max_cstate=9 overrides any DMI blacklist limit. 4678 4679 processor.nocst [HW,ACPI] 4680 Ignore the _CST method to determine C-states, 4681 instead using the legacy FADT method 4682 4683 profile= [KNL] Enable kernel profiling via /proc/profile 4684 Format: [<profiletype>,]<number> 4685 Param: <profiletype>: "schedule" or "kvm" 4686 [defaults to kernel profiling] 4687 Param: "schedule" - profile schedule points. 4688 Param: "kvm" - profile VM exits. 4689 Param: <number> - step/bucket size as a power of 2 for 4690 statistical time based profiling. 4691 4692 prompt_ramdisk= [RAM] [Deprecated] 4693 4694 prot_virt= [S390] enable hosting protected virtual machines 4695 isolated from the hypervisor (if hardware supports 4696 that). 4697 Format: <bool> 4698 4699 psi= [KNL] Enable or disable pressure stall information 4700 tracking. 4701 Format: <bool> 4702 4703 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to 4704 probe for; one of (bare|imps|exps|lifebook|any). 4705 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports 4706 per second. 4707 psmouse.resetafter= [HW,MOUSE] 4708 Try to reset the device after so many bad packets 4709 (0 = never). 4710 psmouse.resolution= 4711 [HW,MOUSE] Set desired mouse resolution, in dpi. 4712 psmouse.smartscroll= 4713 [HW,MOUSE] Controls Logitech smartscroll autorepeat. 4714 0 = disabled, 1 = enabled (default). 4715 4716 pstore.backend= Specify the name of the pstore backend to use 4717 4718 pti= [X86-64] Control Page Table Isolation of user and 4719 kernel address spaces. Disabling this feature 4720 removes hardening, but improves performance of 4721 system calls and interrupts. 4722 4723 on - unconditionally enable 4724 off - unconditionally disable 4725 auto - kernel detects whether your CPU model is 4726 vulnerable to issues that PTI mitigates 4727 4728 Not specifying this option is equivalent to pti=auto. 4729 4730 pty.legacy_count= 4731 [KNL] Number of legacy pty's. Overwrites compiled-in 4732 default number. 4733 4734 quiet [KNL] Disable most log messages 4735 4736 r128= [HW,DRM] 4737 4738 radix_hcall_invalidate=on [PPC/PSERIES] 4739 Disable RADIX GTSE feature and use hcall for TLB 4740 invalidate. 4741 4742 raid= [HW,RAID] 4743 See Documentation/admin-guide/md.rst. 4744 4745 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes 4746 See Documentation/admin-guide/blockdev/ramdisk.rst. 4747 4748 ramdisk_start= [RAM] RAM disk image start address 4749 4750 random.trust_cpu=off 4751 [KNL] Disable trusting the use of the CPU's 4752 random number generator (if available) to 4753 initialize the kernel's RNG. 4754 4755 random.trust_bootloader=off 4756 [KNL] Disable trusting the use of the a seed 4757 passed by the bootloader (if available) to 4758 initialize the kernel's RNG. 4759 4760 randomize_kstack_offset= 4761 [KNL] Enable or disable kernel stack offset 4762 randomization, which provides roughly 5 bits of 4763 entropy, frustrating memory corruption attacks 4764 that depend on stack address determinism or 4765 cross-syscall address exposures. This is only 4766 available on architectures that have defined 4767 CONFIG_HAVE_ARCH_RANDOMIZE_KSTACK_OFFSET. 4768 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 4769 Default is CONFIG_RANDOMIZE_KSTACK_OFFSET_DEFAULT. 4770 4771 ras=option[,option,...] [KNL] RAS-specific options 4772 4773 cec_disable [X86] 4774 Disable the Correctable Errors Collector, 4775 see CONFIG_RAS_CEC help text. 4776 4777 rcu_nocbs[=cpu-list] 4778 [KNL] The optional argument is a cpu list, 4779 as described above. 4780 4781 In kernels built with CONFIG_RCU_NOCB_CPU=y, 4782 enable the no-callback CPU mode, which prevents 4783 such CPUs' callbacks from being invoked in 4784 softirq context. Invocation of such CPUs' RCU 4785 callbacks will instead be offloaded to "rcuox/N" 4786 kthreads created for that purpose, where "x" is 4787 "p" for RCU-preempt, "s" for RCU-sched, and "g" 4788 for the kthreads that mediate grace periods; and 4789 "N" is the CPU number. This reduces OS jitter on 4790 the offloaded CPUs, which can be useful for HPC 4791 and real-time workloads. It can also improve 4792 energy efficiency for asymmetric multiprocessors. 4793 4794 If a cpulist is passed as an argument, the specified 4795 list of CPUs is set to no-callback mode from boot. 4796 4797 Otherwise, if the '=' sign and the cpulist 4798 arguments are omitted, no CPU will be set to 4799 no-callback mode from boot but the mode may be 4800 toggled at runtime via cpusets. 4801 4802 Note that this argument takes precedence over 4803 the CONFIG_RCU_NOCB_CPU_DEFAULT_ALL option. 4804 4805 rcu_nocb_poll [KNL] 4806 Rather than requiring that offloaded CPUs 4807 (specified by rcu_nocbs= above) explicitly 4808 awaken the corresponding "rcuoN" kthreads, 4809 make these kthreads poll for callbacks. 4810 This improves the real-time response for the 4811 offloaded CPUs by relieving them of the need to 4812 wake up the corresponding kthread, but degrades 4813 energy efficiency by requiring that the kthreads 4814 periodically wake up to do the polling. 4815 4816 rcutree.blimit= [KNL] 4817 Set maximum number of finished RCU callbacks to 4818 process in one batch. 4819 4820 rcutree.dump_tree= [KNL] 4821 Dump the structure of the rcu_node combining tree 4822 out at early boot. This is used for diagnostic 4823 purposes, to verify correct tree setup. 4824 4825 rcutree.gp_cleanup_delay= [KNL] 4826 Set the number of jiffies to delay each step of 4827 RCU grace-period cleanup. 4828 4829 rcutree.gp_init_delay= [KNL] 4830 Set the number of jiffies to delay each step of 4831 RCU grace-period initialization. 4832 4833 rcutree.gp_preinit_delay= [KNL] 4834 Set the number of jiffies to delay each step of 4835 RCU grace-period pre-initialization, that is, 4836 the propagation of recent CPU-hotplug changes up 4837 the rcu_node combining tree. 4838 4839 rcutree.jiffies_till_first_fqs= [KNL] 4840 Set delay from grace-period initialization to 4841 first attempt to force quiescent states. 4842 Units are jiffies, minimum value is zero, 4843 and maximum value is HZ. 4844 4845 rcutree.jiffies_till_next_fqs= [KNL] 4846 Set delay between subsequent attempts to force 4847 quiescent states. Units are jiffies, minimum 4848 value is one, and maximum value is HZ. 4849 4850 rcutree.jiffies_till_sched_qs= [KNL] 4851 Set required age in jiffies for a 4852 given grace period before RCU starts 4853 soliciting quiescent-state help from 4854 rcu_note_context_switch() and cond_resched(). 4855 If not specified, the kernel will calculate 4856 a value based on the most recent settings 4857 of rcutree.jiffies_till_first_fqs 4858 and rcutree.jiffies_till_next_fqs. 4859 This calculated value may be viewed in 4860 rcutree.jiffies_to_sched_qs. Any attempt to set 4861 rcutree.jiffies_to_sched_qs will be cheerfully 4862 overwritten. 4863 4864 rcutree.kthread_prio= [KNL,BOOT] 4865 Set the SCHED_FIFO priority of the RCU per-CPU 4866 kthreads (rcuc/N). This value is also used for 4867 the priority of the RCU boost threads (rcub/N) 4868 and for the RCU grace-period kthreads (rcu_bh, 4869 rcu_preempt, and rcu_sched). If RCU_BOOST is 4870 set, valid values are 1-99 and the default is 1 4871 (the least-favored priority). Otherwise, when 4872 RCU_BOOST is not set, valid values are 0-99 and 4873 the default is zero (non-realtime operation). 4874 When RCU_NOCB_CPU is set, also adjust the 4875 priority of NOCB callback kthreads. 4876 4877 rcutree.nocb_nobypass_lim_per_jiffy= [KNL] 4878 On callback-offloaded (rcu_nocbs) CPUs, 4879 RCU reduces the lock contention that would 4880 otherwise be caused by callback floods through 4881 use of the ->nocb_bypass list. However, in the 4882 common non-flooded case, RCU queues directly to 4883 the main ->cblist in order to avoid the extra 4884 overhead of the ->nocb_bypass list and its lock. 4885 But if there are too many callbacks queued during 4886 a single jiffy, RCU pre-queues the callbacks into 4887 the ->nocb_bypass queue. The definition of "too 4888 many" is supplied by this kernel boot parameter. 4889 4890 rcutree.qhimark= [KNL] 4891 Set threshold of queued RCU callbacks beyond which 4892 batch limiting is disabled. 4893 4894 rcutree.qlowmark= [KNL] 4895 Set threshold of queued RCU callbacks below which 4896 batch limiting is re-enabled. 4897 4898 rcutree.qovld= [KNL] 4899 Set threshold of queued RCU callbacks beyond which 4900 RCU's force-quiescent-state scan will aggressively 4901 enlist help from cond_resched() and sched IPIs to 4902 help CPUs more quickly reach quiescent states. 4903 Set to less than zero to make this be set based 4904 on rcutree.qhimark at boot time and to zero to 4905 disable more aggressive help enlistment. 4906 4907 rcutree.rcu_delay_page_cache_fill_msec= [KNL] 4908 Set the page-cache refill delay (in milliseconds) 4909 in response to low-memory conditions. The range 4910 of permitted values is in the range 0:100000. 4911 4912 rcutree.rcu_divisor= [KNL] 4913 Set the shift-right count to use to compute 4914 the callback-invocation batch limit bl from 4915 the number of callbacks queued on this CPU. 4916 The result will be bounded below by the value of 4917 the rcutree.blimit kernel parameter. Every bl 4918 callbacks, the softirq handler will exit in 4919 order to allow the CPU to do other work. 4920 4921 Please note that this callback-invocation batch 4922 limit applies only to non-offloaded callback 4923 invocation. Offloaded callbacks are instead 4924 invoked in the context of an rcuoc kthread, which 4925 scheduler will preempt as it does any other task. 4926 4927 rcutree.rcu_fanout_exact= [KNL] 4928 Disable autobalancing of the rcu_node combining 4929 tree. This is used by rcutorture, and might 4930 possibly be useful for architectures having high 4931 cache-to-cache transfer latencies. 4932 4933 rcutree.rcu_fanout_leaf= [KNL] 4934 Change the number of CPUs assigned to each 4935 leaf rcu_node structure. Useful for very 4936 large systems, which will choose the value 64, 4937 and for NUMA systems with large remote-access 4938 latencies, which will choose a value aligned 4939 with the appropriate hardware boundaries. 4940 4941 rcutree.rcu_min_cached_objs= [KNL] 4942 Minimum number of objects which are cached and 4943 maintained per one CPU. Object size is equal 4944 to PAGE_SIZE. The cache allows to reduce the 4945 pressure to page allocator, also it makes the 4946 whole algorithm to behave better in low memory 4947 condition. 4948 4949 rcutree.rcu_nocb_gp_stride= [KNL] 4950 Set the number of NOCB callback kthreads in 4951 each group, which defaults to the square root 4952 of the number of CPUs. Larger numbers reduce 4953 the wakeup overhead on the global grace-period 4954 kthread, but increases that same overhead on 4955 each group's NOCB grace-period kthread. 4956 4957 rcutree.rcu_kick_kthreads= [KNL] 4958 Cause the grace-period kthread to get an extra 4959 wake_up() if it sleeps three times longer than 4960 it should at force-quiescent-state time. 4961 This wake_up() will be accompanied by a 4962 WARN_ONCE() splat and an ftrace_dump(). 4963 4964 rcutree.rcu_resched_ns= [KNL] 4965 Limit the time spend invoking a batch of RCU 4966 callbacks to the specified number of nanoseconds. 4967 By default, this limit is checked only once 4968 every 32 callbacks in order to limit the pain 4969 inflicted by local_clock() overhead. 4970 4971 rcutree.rcu_unlock_delay= [KNL] 4972 In CONFIG_RCU_STRICT_GRACE_PERIOD=y kernels, 4973 this specifies an rcu_read_unlock()-time delay 4974 in microseconds. This defaults to zero. 4975 Larger delays increase the probability of 4976 catching RCU pointer leaks, that is, buggy use 4977 of RCU-protected pointers after the relevant 4978 rcu_read_unlock() has completed. 4979 4980 rcutree.sysrq_rcu= [KNL] 4981 Commandeer a sysrq key to dump out Tree RCU's 4982 rcu_node tree with an eye towards determining 4983 why a new grace period has not yet started. 4984 4985 rcutree.use_softirq= [KNL] 4986 If set to zero, move all RCU_SOFTIRQ processing to 4987 per-CPU rcuc kthreads. Defaults to a non-zero 4988 value, meaning that RCU_SOFTIRQ is used by default. 4989 Specify rcutree.use_softirq=0 to use rcuc kthreads. 4990 4991 But note that CONFIG_PREEMPT_RT=y kernels disable 4992 this kernel boot parameter, forcibly setting it 4993 to zero. 4994 4995 rcuscale.gp_async= [KNL] 4996 Measure performance of asynchronous 4997 grace-period primitives such as call_rcu(). 4998 4999 rcuscale.gp_async_max= [KNL] 5000 Specify the maximum number of outstanding 5001 callbacks per writer thread. When a writer 5002 thread exceeds this limit, it invokes the 5003 corresponding flavor of rcu_barrier() to allow 5004 previously posted callbacks to drain. 5005 5006 rcuscale.gp_exp= [KNL] 5007 Measure performance of expedited synchronous 5008 grace-period primitives. 5009 5010 rcuscale.holdoff= [KNL] 5011 Set test-start holdoff period. The purpose of 5012 this parameter is to delay the start of the 5013 test until boot completes in order to avoid 5014 interference. 5015 5016 rcuscale.kfree_by_call_rcu= [KNL] 5017 In kernels built with CONFIG_RCU_LAZY=y, test 5018 call_rcu() instead of kfree_rcu(). 5019 5020 rcuscale.kfree_mult= [KNL] 5021 Instead of allocating an object of size kfree_obj, 5022 allocate one of kfree_mult * sizeof(kfree_obj). 5023 Defaults to 1. 5024 5025 rcuscale.kfree_rcu_test= [KNL] 5026 Set to measure performance of kfree_rcu() flooding. 5027 5028 rcuscale.kfree_rcu_test_double= [KNL] 5029 Test the double-argument variant of kfree_rcu(). 5030 If this parameter has the same value as 5031 rcuscale.kfree_rcu_test_single, both the single- 5032 and double-argument variants are tested. 5033 5034 rcuscale.kfree_rcu_test_single= [KNL] 5035 Test the single-argument variant of kfree_rcu(). 5036 If this parameter has the same value as 5037 rcuscale.kfree_rcu_test_double, both the single- 5038 and double-argument variants are tested. 5039 5040 rcuscale.kfree_nthreads= [KNL] 5041 The number of threads running loops of kfree_rcu(). 5042 5043 rcuscale.kfree_alloc_num= [KNL] 5044 Number of allocations and frees done in an iteration. 5045 5046 rcuscale.kfree_loops= [KNL] 5047 Number of loops doing rcuscale.kfree_alloc_num number 5048 of allocations and frees. 5049 5050 rcuscale.minruntime= [KNL] 5051 Set the minimum test run time in seconds. This 5052 does not affect the data-collection interval, 5053 but instead allows better measurement of things 5054 like CPU consumption. 5055 5056 rcuscale.nreaders= [KNL] 5057 Set number of RCU readers. The value -1 selects 5058 N, where N is the number of CPUs. A value 5059 "n" less than -1 selects N-n+1, where N is again 5060 the number of CPUs. For example, -2 selects N 5061 (the number of CPUs), -3 selects N+1, and so on. 5062 A value of "n" less than or equal to -N selects 5063 a single reader. 5064 5065 rcuscale.nwriters= [KNL] 5066 Set number of RCU writers. The values operate 5067 the same as for rcuscale.nreaders. 5068 N, where N is the number of CPUs 5069 5070 rcuscale.scale_type= [KNL] 5071 Specify the RCU implementation to test. 5072 5073 rcuscale.shutdown= [KNL] 5074 Shut the system down after performance tests 5075 complete. This is useful for hands-off automated 5076 testing. 5077 5078 rcuscale.verbose= [KNL] 5079 Enable additional printk() statements. 5080 5081 rcuscale.writer_holdoff= [KNL] 5082 Write-side holdoff between grace periods, 5083 in microseconds. The default of zero says 5084 no holdoff. 5085 5086 rcuscale.writer_holdoff_jiffies= [KNL] 5087 Additional write-side holdoff between grace 5088 periods, but in jiffies. The default of zero 5089 says no holdoff. 5090 5091 rcutorture.fqs_duration= [KNL] 5092 Set duration of force_quiescent_state bursts 5093 in microseconds. 5094 5095 rcutorture.fqs_holdoff= [KNL] 5096 Set holdoff time within force_quiescent_state bursts 5097 in microseconds. 5098 5099 rcutorture.fqs_stutter= [KNL] 5100 Set wait time between force_quiescent_state bursts 5101 in seconds. 5102 5103 rcutorture.fwd_progress= [KNL] 5104 Specifies the number of kthreads to be used 5105 for RCU grace-period forward-progress testing 5106 for the types of RCU supporting this notion. 5107 Defaults to 1 kthread, values less than zero or 5108 greater than the number of CPUs cause the number 5109 of CPUs to be used. 5110 5111 rcutorture.fwd_progress_div= [KNL] 5112 Specify the fraction of a CPU-stall-warning 5113 period to do tight-loop forward-progress testing. 5114 5115 rcutorture.fwd_progress_holdoff= [KNL] 5116 Number of seconds to wait between successive 5117 forward-progress tests. 5118 5119 rcutorture.fwd_progress_need_resched= [KNL] 5120 Enclose cond_resched() calls within checks for 5121 need_resched() during tight-loop forward-progress 5122 testing. 5123 5124 rcutorture.gp_cond= [KNL] 5125 Use conditional/asynchronous update-side 5126 primitives, if available. 5127 5128 rcutorture.gp_exp= [KNL] 5129 Use expedited update-side primitives, if available. 5130 5131 rcutorture.gp_normal= [KNL] 5132 Use normal (non-expedited) asynchronous 5133 update-side primitives, if available. 5134 5135 rcutorture.gp_sync= [KNL] 5136 Use normal (non-expedited) synchronous 5137 update-side primitives, if available. If all 5138 of rcutorture.gp_cond=, rcutorture.gp_exp=, 5139 rcutorture.gp_normal=, and rcutorture.gp_sync= 5140 are zero, rcutorture acts as if is interpreted 5141 they are all non-zero. 5142 5143 rcutorture.irqreader= [KNL] 5144 Run RCU readers from irq handlers, or, more 5145 accurately, from a timer handler. Not all RCU 5146 flavors take kindly to this sort of thing. 5147 5148 rcutorture.leakpointer= [KNL] 5149 Leak an RCU-protected pointer out of the reader. 5150 This can of course result in splats, and is 5151 intended to test the ability of things like 5152 CONFIG_RCU_STRICT_GRACE_PERIOD=y to detect 5153 such leaks. 5154 5155 rcutorture.n_barrier_cbs= [KNL] 5156 Set callbacks/threads for rcu_barrier() testing. 5157 5158 rcutorture.nfakewriters= [KNL] 5159 Set number of concurrent RCU writers. These just 5160 stress RCU, they don't participate in the actual 5161 test, hence the "fake". 5162 5163 rcutorture.nocbs_nthreads= [KNL] 5164 Set number of RCU callback-offload togglers. 5165 Zero (the default) disables toggling. 5166 5167 rcutorture.nocbs_toggle= [KNL] 5168 Set the delay in milliseconds between successive 5169 callback-offload toggling attempts. 5170 5171 rcutorture.nreaders= [KNL] 5172 Set number of RCU readers. The value -1 selects 5173 N-1, where N is the number of CPUs. A value 5174 "n" less than -1 selects N-n-2, where N is again 5175 the number of CPUs. For example, -2 selects N 5176 (the number of CPUs), -3 selects N+1, and so on. 5177 5178 rcutorture.object_debug= [KNL] 5179 Enable debug-object double-call_rcu() testing. 5180 5181 rcutorture.onoff_holdoff= [KNL] 5182 Set time (s) after boot for CPU-hotplug testing. 5183 5184 rcutorture.onoff_interval= [KNL] 5185 Set time (jiffies) between CPU-hotplug operations, 5186 or zero to disable CPU-hotplug testing. 5187 5188 rcutorture.read_exit= [KNL] 5189 Set the number of read-then-exit kthreads used 5190 to test the interaction of RCU updaters and 5191 task-exit processing. 5192 5193 rcutorture.read_exit_burst= [KNL] 5194 The number of times in a given read-then-exit 5195 episode that a set of read-then-exit kthreads 5196 is spawned. 5197 5198 rcutorture.read_exit_delay= [KNL] 5199 The delay, in seconds, between successive 5200 read-then-exit testing episodes. 5201 5202 rcutorture.shuffle_interval= [KNL] 5203 Set task-shuffle interval (s). Shuffling tasks 5204 allows some CPUs to go into dyntick-idle mode 5205 during the rcutorture test. 5206 5207 rcutorture.shutdown_secs= [KNL] 5208 Set time (s) after boot system shutdown. This 5209 is useful for hands-off automated testing. 5210 5211 rcutorture.stall_cpu= [KNL] 5212 Duration of CPU stall (s) to test RCU CPU stall 5213 warnings, zero to disable. 5214 5215 rcutorture.stall_cpu_block= [KNL] 5216 Sleep while stalling if set. This will result 5217 in warnings from preemptible RCU in addition to 5218 any other stall-related activity. Note that 5219 in kernels built with CONFIG_PREEMPTION=n and 5220 CONFIG_PREEMPT_COUNT=y, this parameter will 5221 cause the CPU to pass through a quiescent state. 5222 Given CONFIG_PREEMPTION=n, this will suppress 5223 RCU CPU stall warnings, but will instead result 5224 in scheduling-while-atomic splats. 5225 5226 Use of this module parameter results in splats. 5227 5228 5229 rcutorture.stall_cpu_holdoff= [KNL] 5230 Time to wait (s) after boot before inducing stall. 5231 5232 rcutorture.stall_cpu_irqsoff= [KNL] 5233 Disable interrupts while stalling if set. 5234 5235 rcutorture.stall_gp_kthread= [KNL] 5236 Duration (s) of forced sleep within RCU 5237 grace-period kthread to test RCU CPU stall 5238 warnings, zero to disable. If both stall_cpu 5239 and stall_gp_kthread are specified, the 5240 kthread is starved first, then the CPU. 5241 5242 rcutorture.stat_interval= [KNL] 5243 Time (s) between statistics printk()s. 5244 5245 rcutorture.stutter= [KNL] 5246 Time (s) to stutter testing, for example, specifying 5247 five seconds causes the test to run for five seconds, 5248 wait for five seconds, and so on. This tests RCU's 5249 ability to transition abruptly to and from idle. 5250 5251 rcutorture.test_boost= [KNL] 5252 Test RCU priority boosting? 0=no, 1=maybe, 2=yes. 5253 "Maybe" means test if the RCU implementation 5254 under test support RCU priority boosting. 5255 5256 rcutorture.test_boost_duration= [KNL] 5257 Duration (s) of each individual boost test. 5258 5259 rcutorture.test_boost_interval= [KNL] 5260 Interval (s) between each boost test. 5261 5262 rcutorture.test_no_idle_hz= [KNL] 5263 Test RCU's dyntick-idle handling. See also the 5264 rcutorture.shuffle_interval parameter. 5265 5266 rcutorture.torture_type= [KNL] 5267 Specify the RCU implementation to test. 5268 5269 rcutorture.verbose= [KNL] 5270 Enable additional printk() statements. 5271 5272 rcupdate.rcu_cpu_stall_ftrace_dump= [KNL] 5273 Dump ftrace buffer after reporting RCU CPU 5274 stall warning. 5275 5276 rcupdate.rcu_cpu_stall_suppress= [KNL] 5277 Suppress RCU CPU stall warning messages. 5278 5279 rcupdate.rcu_cpu_stall_suppress_at_boot= [KNL] 5280 Suppress RCU CPU stall warning messages and 5281 rcutorture writer stall warnings that occur 5282 during early boot, that is, during the time 5283 before the init task is spawned. 5284 5285 rcupdate.rcu_cpu_stall_timeout= [KNL] 5286 Set timeout for RCU CPU stall warning messages. 5287 The value is in seconds and the maximum allowed 5288 value is 300 seconds. 5289 5290 rcupdate.rcu_exp_cpu_stall_timeout= [KNL] 5291 Set timeout for expedited RCU CPU stall warning 5292 messages. The value is in milliseconds 5293 and the maximum allowed value is 21000 5294 milliseconds. Please note that this value is 5295 adjusted to an arch timer tick resolution. 5296 Setting this to zero causes the value from 5297 rcupdate.rcu_cpu_stall_timeout to be used (after 5298 conversion from seconds to milliseconds). 5299 5300 rcupdate.rcu_cpu_stall_cputime= [KNL] 5301 Provide statistics on the cputime and count of 5302 interrupts and tasks during the sampling period. For 5303 multiple continuous RCU stalls, all sampling periods 5304 begin at half of the first RCU stall timeout. 5305 5306 rcupdate.rcu_exp_stall_task_details= [KNL] 5307 Print stack dumps of any tasks blocking the 5308 current expedited RCU grace period during an 5309 expedited RCU CPU stall warning. 5310 5311 rcupdate.rcu_expedited= [KNL] 5312 Use expedited grace-period primitives, for 5313 example, synchronize_rcu_expedited() instead 5314 of synchronize_rcu(). This reduces latency, 5315 but can increase CPU utilization, degrade 5316 real-time latency, and degrade energy efficiency. 5317 No effect on CONFIG_TINY_RCU kernels. 5318 5319 rcupdate.rcu_normal= [KNL] 5320 Use only normal grace-period primitives, 5321 for example, synchronize_rcu() instead of 5322 synchronize_rcu_expedited(). This improves 5323 real-time latency, CPU utilization, and 5324 energy efficiency, but can expose users to 5325 increased grace-period latency. This parameter 5326 overrides rcupdate.rcu_expedited. No effect on 5327 CONFIG_TINY_RCU kernels. 5328 5329 rcupdate.rcu_normal_after_boot= [KNL] 5330 Once boot has completed (that is, after 5331 rcu_end_inkernel_boot() has been invoked), use 5332 only normal grace-period primitives. No effect 5333 on CONFIG_TINY_RCU kernels. 5334 5335 But note that CONFIG_PREEMPT_RT=y kernels enables 5336 this kernel boot parameter, forcibly setting 5337 it to the value one, that is, converting any 5338 post-boot attempt at an expedited RCU grace 5339 period to instead use normal non-expedited 5340 grace-period processing. 5341 5342 rcupdate.rcu_task_collapse_lim= [KNL] 5343 Set the maximum number of callbacks present 5344 at the beginning of a grace period that allows 5345 the RCU Tasks flavors to collapse back to using 5346 a single callback queue. This switching only 5347 occurs when rcupdate.rcu_task_enqueue_lim is 5348 set to the default value of -1. 5349 5350 rcupdate.rcu_task_contend_lim= [KNL] 5351 Set the minimum number of callback-queuing-time 5352 lock-contention events per jiffy required to 5353 cause the RCU Tasks flavors to switch to per-CPU 5354 callback queuing. This switching only occurs 5355 when rcupdate.rcu_task_enqueue_lim is set to 5356 the default value of -1. 5357 5358 rcupdate.rcu_task_enqueue_lim= [KNL] 5359 Set the number of callback queues to use for the 5360 RCU Tasks family of RCU flavors. The default 5361 of -1 allows this to be automatically (and 5362 dynamically) adjusted. This parameter is intended 5363 for use in testing. 5364 5365 rcupdate.rcu_task_ipi_delay= [KNL] 5366 Set time in jiffies during which RCU tasks will 5367 avoid sending IPIs, starting with the beginning 5368 of a given grace period. Setting a large 5369 number avoids disturbing real-time workloads, 5370 but lengthens grace periods. 5371 5372 rcupdate.rcu_task_lazy_lim= [KNL] 5373 Number of callbacks on a given CPU that will 5374 cancel laziness on that CPU. Use -1 to disable 5375 cancellation of laziness, but be advised that 5376 doing so increases the danger of OOM due to 5377 callback flooding. 5378 5379 rcupdate.rcu_task_stall_info= [KNL] 5380 Set initial timeout in jiffies for RCU task stall 5381 informational messages, which give some indication 5382 of the problem for those not patient enough to 5383 wait for ten minutes. Informational messages are 5384 only printed prior to the stall-warning message 5385 for a given grace period. Disable with a value 5386 less than or equal to zero. Defaults to ten 5387 seconds. A change in value does not take effect 5388 until the beginning of the next grace period. 5389 5390 rcupdate.rcu_task_stall_info_mult= [KNL] 5391 Multiplier for time interval between successive 5392 RCU task stall informational messages for a given 5393 RCU tasks grace period. This value is clamped 5394 to one through ten, inclusive. It defaults to 5395 the value three, so that the first informational 5396 message is printed 10 seconds into the grace 5397 period, the second at 40 seconds, the third at 5398 160 seconds, and then the stall warning at 600 5399 seconds would prevent a fourth at 640 seconds. 5400 5401 rcupdate.rcu_task_stall_timeout= [KNL] 5402 Set timeout in jiffies for RCU task stall 5403 warning messages. Disable with a value less 5404 than or equal to zero. Defaults to ten minutes. 5405 A change in value does not take effect until 5406 the beginning of the next grace period. 5407 5408 rcupdate.rcu_tasks_lazy_ms= [KNL] 5409 Set timeout in milliseconds RCU Tasks asynchronous 5410 callback batching for call_rcu_tasks(). 5411 A negative value will take the default. A value 5412 of zero will disable batching. Batching is 5413 always disabled for synchronize_rcu_tasks(). 5414 5415 rcupdate.rcu_tasks_rude_lazy_ms= [KNL] 5416 Set timeout in milliseconds RCU Tasks 5417 Rude asynchronous callback batching for 5418 call_rcu_tasks_rude(). A negative value 5419 will take the default. A value of zero will 5420 disable batching. Batching is always disabled 5421 for synchronize_rcu_tasks_rude(). 5422 5423 rcupdate.rcu_tasks_trace_lazy_ms= [KNL] 5424 Set timeout in milliseconds RCU Tasks 5425 Trace asynchronous callback batching for 5426 call_rcu_tasks_trace(). A negative value 5427 will take the default. A value of zero will 5428 disable batching. Batching is always disabled 5429 for synchronize_rcu_tasks_trace(). 5430 5431 rcupdate.rcu_self_test= [KNL] 5432 Run the RCU early boot self tests 5433 5434 rdinit= [KNL] 5435 Format: <full_path> 5436 Run specified binary instead of /init from the ramdisk, 5437 used for early userspace startup. See initrd. 5438 5439 rdrand= [X86] 5440 force - Override the decision by the kernel to hide the 5441 advertisement of RDRAND support (this affects 5442 certain AMD processors because of buggy BIOS 5443 support, specifically around the suspend/resume 5444 path). 5445 5446 rdt= [HW,X86,RDT] 5447 Turn on/off individual RDT features. List is: 5448 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp, 5449 mba, smba, bmec. 5450 E.g. to turn on cmt and turn off mba use: 5451 rdt=cmt,!mba 5452 5453 reboot= [KNL] 5454 Format (x86 or x86_64): 5455 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] | d[efault] \ 5456 [[,]s[mp]#### \ 5457 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \ 5458 [[,]f[orce] 5459 Where reboot_mode is one of warm (soft) or cold (hard) or gpio 5460 (prefix with 'panic_' to set mode for panic 5461 reboot only), 5462 reboot_type is one of bios, acpi, kbd, triple, efi, or pci, 5463 reboot_force is either force or not specified, 5464 reboot_cpu is s[mp]#### with #### being the processor 5465 to be used for rebooting. 5466 5467 refscale.holdoff= [KNL] 5468 Set test-start holdoff period. The purpose of 5469 this parameter is to delay the start of the 5470 test until boot completes in order to avoid 5471 interference. 5472 5473 refscale.loops= [KNL] 5474 Set the number of loops over the synchronization 5475 primitive under test. Increasing this number 5476 reduces noise due to loop start/end overhead, 5477 but the default has already reduced the per-pass 5478 noise to a handful of picoseconds on ca. 2020 5479 x86 laptops. 5480 5481 refscale.nreaders= [KNL] 5482 Set number of readers. The default value of -1 5483 selects N, where N is roughly 75% of the number 5484 of CPUs. A value of zero is an interesting choice. 5485 5486 refscale.nruns= [KNL] 5487 Set number of runs, each of which is dumped onto 5488 the console log. 5489 5490 refscale.readdelay= [KNL] 5491 Set the read-side critical-section duration, 5492 measured in microseconds. 5493 5494 refscale.scale_type= [KNL] 5495 Specify the read-protection implementation to test. 5496 5497 refscale.shutdown= [KNL] 5498 Shut down the system at the end of the performance 5499 test. This defaults to 1 (shut it down) when 5500 refscale is built into the kernel and to 0 (leave 5501 it running) when refscale is built as a module. 5502 5503 refscale.verbose= [KNL] 5504 Enable additional printk() statements. 5505 5506 refscale.verbose_batched= [KNL] 5507 Batch the additional printk() statements. If zero 5508 (the default) or negative, print everything. Otherwise, 5509 print every Nth verbose statement, where N is the value 5510 specified. 5511 5512 relax_domain_level= 5513 [KNL, SMP] Set scheduler's default relax_domain_level. 5514 See Documentation/admin-guide/cgroup-v1/cpusets.rst. 5515 5516 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory 5517 Format: <base1>,<size1>[,<base2>,<size2>,...] 5518 Reserve I/O ports or memory so the kernel won't use 5519 them. If <base> is less than 0x10000, the region 5520 is assumed to be I/O ports; otherwise it is memory. 5521 5522 reservetop= [X86-32] 5523 Format: nn[KMG] 5524 Reserves a hole at the top of the kernel virtual 5525 address space. 5526 5527 reset_devices [KNL] Force drivers to reset the underlying device 5528 during initialization. 5529 5530 resume= [SWSUSP] 5531 Specify the partition device for software suspend 5532 Format: 5533 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>} 5534 5535 resume_offset= [SWSUSP] 5536 Specify the offset from the beginning of the partition 5537 given by "resume=" at which the swap header is located, 5538 in <PAGE_SIZE> units (needed only for swap files). 5539 See Documentation/power/swsusp-and-swap-files.rst 5540 5541 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to 5542 read the resume files 5543 5544 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up. 5545 Useful for devices that are detected asynchronously 5546 (e.g. USB and MMC devices). 5547 5548 retain_initrd [RAM] Keep initrd memory after extraction 5549 5550 retbleed= [X86] Control mitigation of RETBleed (Arbitrary 5551 Speculative Code Execution with Return Instructions) 5552 vulnerability. 5553 5554 AMD-based UNRET and IBPB mitigations alone do not stop 5555 sibling threads from influencing the predictions of other 5556 sibling threads. For that reason, STIBP is used on pro- 5557 cessors that support it, and mitigate SMT on processors 5558 that don't. 5559 5560 off - no mitigation 5561 auto - automatically select a migitation 5562 auto,nosmt - automatically select a mitigation, 5563 disabling SMT if necessary for 5564 the full mitigation (only on Zen1 5565 and older without STIBP). 5566 ibpb - On AMD, mitigate short speculation 5567 windows on basic block boundaries too. 5568 Safe, highest perf impact. It also 5569 enables STIBP if present. Not suitable 5570 on Intel. 5571 ibpb,nosmt - Like "ibpb" above but will disable SMT 5572 when STIBP is not available. This is 5573 the alternative for systems which do not 5574 have STIBP. 5575 unret - Force enable untrained return thunks, 5576 only effective on AMD f15h-f17h based 5577 systems. 5578 unret,nosmt - Like unret, but will disable SMT when STIBP 5579 is not available. This is the alternative for 5580 systems which do not have STIBP. 5581 5582 Selecting 'auto' will choose a mitigation method at run 5583 time according to the CPU. 5584 5585 Not specifying this option is equivalent to retbleed=auto. 5586 5587 rfkill.default_state= 5588 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm, 5589 etc. communication is blocked by default. 5590 1 Unblocked. 5591 5592 rfkill.master_switch_mode= 5593 0 The "airplane mode" button does nothing. 5594 1 The "airplane mode" button toggles between everything 5595 blocked and the previous configuration. 5596 2 The "airplane mode" button toggles between everything 5597 blocked and everything unblocked. 5598 5599 rhash_entries= [KNL,NET] 5600 Set number of hash buckets for route cache 5601 5602 ring3mwait=disable 5603 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported 5604 CPUs. 5605 5606 riscv_isa_fallback [RISCV] 5607 When CONFIG_RISCV_ISA_FALLBACK is not enabled, permit 5608 falling back to detecting extension support by parsing 5609 "riscv,isa" property on devicetree systems when the 5610 replacement properties are not found. See the Kconfig 5611 entry for RISCV_ISA_FALLBACK. 5612 5613 ro [KNL] Mount root device read-only on boot 5614 5615 rodata= [KNL] 5616 on Mark read-only kernel memory as read-only (default). 5617 off Leave read-only kernel memory writable for debugging. 5618 full Mark read-only kernel memory and aliases as read-only 5619 [arm64] 5620 5621 rockchip.usb_uart 5622 Enable the uart passthrough on the designated usb port 5623 on Rockchip SoCs. When active, the signals of the 5624 debug-uart get routed to the D+ and D- pins of the usb 5625 port and the regular usb controller gets disabled. 5626 5627 root= [KNL] Root filesystem 5628 Usually this a a block device specifier of some kind, 5629 see the early_lookup_bdev comment in 5630 block/early-lookup.c for details. 5631 Alternatively this can be "ram" for the legacy initial 5632 ramdisk, "nfs" and "cifs" for root on a network file 5633 system, or "mtd" and "ubi" for mounting from raw flash. 5634 5635 rootdelay= [KNL] Delay (in seconds) to pause before attempting to 5636 mount the root filesystem 5637 5638 rootflags= [KNL] Set root filesystem mount option string 5639 5640 rootfstype= [KNL] Set root filesystem type 5641 5642 rootwait [KNL] Wait (indefinitely) for root device to show up. 5643 Useful for devices that are detected asynchronously 5644 (e.g. USB and MMC devices). 5645 5646 rootwait= [KNL] Maximum time (in seconds) to wait for root device 5647 to show up before attempting to mount the root 5648 filesystem. 5649 5650 rproc_mem=nn[KMG][@address] 5651 [KNL,ARM,CMA] Remoteproc physical memory block. 5652 Memory area to be used by remote processor image, 5653 managed by CMA. 5654 5655 rw [KNL] Mount root device read-write on boot 5656 5657 S [KNL] Run init in single mode 5658 5659 s390_iommu= [HW,S390] 5660 Set s390 IOTLB flushing mode 5661 strict 5662 With strict flushing every unmap operation will result in 5663 an IOTLB flush. Default is lazy flushing before reuse, 5664 which is faster. 5665 5666 s390_iommu_aperture= [KNL,S390] 5667 Specifies the size of the per device DMA address space 5668 accessible through the DMA and IOMMU APIs as a decimal 5669 factor of the size of main memory. 5670 The default is 1 meaning that one can concurrently use 5671 as many DMA addresses as physical memory is installed, 5672 if supported by hardware, and thus map all of memory 5673 once. With a value of 2 one can map all of memory twice 5674 and so on. As a special case a factor of 0 imposes no 5675 restrictions other than those given by hardware at the 5676 cost of significant additional memory use for tables. 5677 5678 sa1100ir [NET] 5679 See drivers/net/irda/sa1100_ir.c. 5680 5681 sched_verbose [KNL] Enables verbose scheduler debug messages. 5682 5683 schedstats= [KNL,X86] Enable or disable scheduled statistics. 5684 Allowed values are enable and disable. This feature 5685 incurs a small amount of overhead in the scheduler 5686 but is useful for debugging and performance tuning. 5687 5688 sched_thermal_decay_shift= 5689 [KNL, SMP] Set a decay shift for scheduler thermal 5690 pressure signal. Thermal pressure signal follows the 5691 default decay period of other scheduler pelt 5692 signals(usually 32 ms but configurable). Setting 5693 sched_thermal_decay_shift will left shift the decay 5694 period for the thermal pressure signal by the shift 5695 value. 5696 i.e. with the default pelt decay period of 32 ms 5697 sched_thermal_decay_shift thermal pressure decay pr 5698 1 64 ms 5699 2 128 ms 5700 and so on. 5701 Format: integer between 0 and 10 5702 Default is 0. 5703 5704 scftorture.holdoff= [KNL] 5705 Number of seconds to hold off before starting 5706 test. Defaults to zero for module insertion and 5707 to 10 seconds for built-in smp_call_function() 5708 tests. 5709 5710 scftorture.longwait= [KNL] 5711 Request ridiculously long waits randomly selected 5712 up to the chosen limit in seconds. Zero (the 5713 default) disables this feature. Please note 5714 that requesting even small non-zero numbers of 5715 seconds can result in RCU CPU stall warnings, 5716 softlockup complaints, and so on. 5717 5718 scftorture.nthreads= [KNL] 5719 Number of kthreads to spawn to invoke the 5720 smp_call_function() family of functions. 5721 The default of -1 specifies a number of kthreads 5722 equal to the number of CPUs. 5723 5724 scftorture.onoff_holdoff= [KNL] 5725 Number seconds to wait after the start of the 5726 test before initiating CPU-hotplug operations. 5727 5728 scftorture.onoff_interval= [KNL] 5729 Number seconds to wait between successive 5730 CPU-hotplug operations. Specifying zero (which 5731 is the default) disables CPU-hotplug operations. 5732 5733 scftorture.shutdown_secs= [KNL] 5734 The number of seconds following the start of the 5735 test after which to shut down the system. The 5736 default of zero avoids shutting down the system. 5737 Non-zero values are useful for automated tests. 5738 5739 scftorture.stat_interval= [KNL] 5740 The number of seconds between outputting the 5741 current test statistics to the console. A value 5742 of zero disables statistics output. 5743 5744 scftorture.stutter_cpus= [KNL] 5745 The number of jiffies to wait between each change 5746 to the set of CPUs under test. 5747 5748 scftorture.use_cpus_read_lock= [KNL] 5749 Use use_cpus_read_lock() instead of the default 5750 preempt_disable() to disable CPU hotplug 5751 while invoking one of the smp_call_function*() 5752 functions. 5753 5754 scftorture.verbose= [KNL] 5755 Enable additional printk() statements. 5756 5757 scftorture.weight_single= [KNL] 5758 The probability weighting to use for the 5759 smp_call_function_single() function with a zero 5760 "wait" parameter. A value of -1 selects the 5761 default if all other weights are -1. However, 5762 if at least one weight has some other value, a 5763 value of -1 will instead select a weight of zero. 5764 5765 scftorture.weight_single_wait= [KNL] 5766 The probability weighting to use for the 5767 smp_call_function_single() function with a 5768 non-zero "wait" parameter. See weight_single. 5769 5770 scftorture.weight_many= [KNL] 5771 The probability weighting to use for the 5772 smp_call_function_many() function with a zero 5773 "wait" parameter. See weight_single. 5774 Note well that setting a high probability for 5775 this weighting can place serious IPI load 5776 on the system. 5777 5778 scftorture.weight_many_wait= [KNL] 5779 The probability weighting to use for the 5780 smp_call_function_many() function with a 5781 non-zero "wait" parameter. See weight_single 5782 and weight_many. 5783 5784 scftorture.weight_all= [KNL] 5785 The probability weighting to use for the 5786 smp_call_function_all() function with a zero 5787 "wait" parameter. See weight_single and 5788 weight_many. 5789 5790 scftorture.weight_all_wait= [KNL] 5791 The probability weighting to use for the 5792 smp_call_function_all() function with a 5793 non-zero "wait" parameter. See weight_single 5794 and weight_many. 5795 5796 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate 5797 xtime_lock contention on larger systems, and/or RCU lock 5798 contention on all systems with CONFIG_MAXSMP set. 5799 Format: { "0" | "1" } 5800 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1" 5801 1 -- enable. 5802 Note: increases power consumption, thus should only be 5803 enabled if running jitter sensitive (HPC/RT) workloads. 5804 5805 security= [SECURITY] Choose a legacy "major" security module to 5806 enable at boot. This has been deprecated by the 5807 "lsm=" parameter. 5808 5809 selinux= [SELINUX] Disable or enable SELinux at boot time. 5810 Format: { "0" | "1" } 5811 See security/selinux/Kconfig help text. 5812 0 -- disable. 5813 1 -- enable. 5814 Default value is 1. 5815 5816 serialnumber [BUGS=X86-32] 5817 5818 sev=option[,option...] [X86-64] See Documentation/arch/x86/x86_64/boot-options.rst 5819 5820 shapers= [NET] 5821 Maximal number of shapers. 5822 5823 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller 5824 Limit apic dumping. The parameter defines the maximal 5825 number of local apics being dumped. Also it is possible 5826 to set it to "all" by meaning -- no limit here. 5827 Format: { 1 (default) | 2 | ... | all }. 5828 The parameter valid if only apic=debug or 5829 apic=verbose is specified. 5830 Example: apic=debug show_lapic=all 5831 5832 simeth= [IA-64] 5833 simscsi= 5834 5835 slram= [HW,MTD] 5836 5837 slab_merge [MM] 5838 Enable merging of slabs with similar size when the 5839 kernel is built without CONFIG_SLAB_MERGE_DEFAULT. 5840 5841 slab_nomerge [MM] 5842 Disable merging of slabs with similar size. May be 5843 necessary if there is some reason to distinguish 5844 allocs to different slabs, especially in hardened 5845 environments where the risk of heap overflows and 5846 layout control by attackers can usually be 5847 frustrated by disabling merging. This will reduce 5848 most of the exposure of a heap attack to a single 5849 cache (risks via metadata attacks are mostly 5850 unchanged). Debug options disable merging on their 5851 own. 5852 For more information see Documentation/mm/slub.rst. 5853 5854 slab_max_order= [MM, SLAB] 5855 Determines the maximum allowed order for slabs. 5856 A high setting may cause OOMs due to memory 5857 fragmentation. Defaults to 1 for systems with 5858 more than 32MB of RAM, 0 otherwise. 5859 5860 slub_debug[=options[,slabs][;[options[,slabs]]...] [MM, SLUB] 5861 Enabling slub_debug allows one to determine the 5862 culprit if slab objects become corrupted. Enabling 5863 slub_debug can create guard zones around objects and 5864 may poison objects when not in use. Also tracks the 5865 last alloc / free. For more information see 5866 Documentation/mm/slub.rst. 5867 5868 slub_max_order= [MM, SLUB] 5869 Determines the maximum allowed order for slabs. 5870 A high setting may cause OOMs due to memory 5871 fragmentation. For more information see 5872 Documentation/mm/slub.rst. 5873 5874 slub_min_objects= [MM, SLUB] 5875 The minimum number of objects per slab. SLUB will 5876 increase the slab order up to slub_max_order to 5877 generate a sufficiently large slab able to contain 5878 the number of objects indicated. The higher the number 5879 of objects the smaller the overhead of tracking slabs 5880 and the less frequently locks need to be acquired. 5881 For more information see Documentation/mm/slub.rst. 5882 5883 slub_min_order= [MM, SLUB] 5884 Determines the minimum page order for slabs. Must be 5885 lower than slub_max_order. 5886 For more information see Documentation/mm/slub.rst. 5887 5888 slub_merge [MM, SLUB] 5889 Same with slab_merge. 5890 5891 slub_nomerge [MM, SLUB] 5892 Same with slab_nomerge. This is supported for legacy. 5893 See slab_nomerge for more information. 5894 5895 smart2= [HW] 5896 Format: <io1>[,<io2>[,...,<io8>]] 5897 5898 smp.csd_lock_timeout= [KNL] 5899 Specify the period of time in milliseconds 5900 that smp_call_function() and friends will wait 5901 for a CPU to release the CSD lock. This is 5902 useful when diagnosing bugs involving CPUs 5903 disabling interrupts for extended periods 5904 of time. Defaults to 5,000 milliseconds, and 5905 setting a value of zero disables this feature. 5906 This feature may be more efficiently disabled 5907 using the csdlock_debug- kernel parameter. 5908 5909 smp.panic_on_ipistall= [KNL] 5910 If a csd_lock_timeout extends for more than 5911 the specified number of milliseconds, panic the 5912 system. By default, let CSD-lock acquisition 5913 take as long as they take. Specifying 300,000 5914 for this value provides a 5-minute timeout. 5915 5916 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices 5917 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port 5918 smsc-ircc2.ircc_sir= [HW] SIR base I/O port 5919 smsc-ircc2.ircc_fir= [HW] FIR base I/O port 5920 smsc-ircc2.ircc_irq= [HW] IRQ line 5921 smsc-ircc2.ircc_dma= [HW] DMA channel 5922 smsc-ircc2.ircc_transceiver= [HW] Transceiver type: 5923 0: Toshiba Satellite 1800 (GP data pin select) 5924 1: Fast pin select (default) 5925 2: ATC IRMode 5926 5927 smt= [KNL,MIPS,S390] Set the maximum number of threads (logical 5928 CPUs) to use per physical CPU on systems capable of 5929 symmetric multithreading (SMT). Will be capped to the 5930 actual hardware limit. 5931 Format: <integer> 5932 Default: -1 (no limit) 5933 5934 softlockup_panic= 5935 [KNL] Should the soft-lockup detector generate panics. 5936 Format: 0 | 1 5937 5938 A value of 1 instructs the soft-lockup detector 5939 to panic the machine when a soft-lockup occurs. It is 5940 also controlled by the kernel.softlockup_panic sysctl 5941 and CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC, which is the 5942 respective build-time switch to that functionality. 5943 5944 softlockup_all_cpu_backtrace= 5945 [KNL] Should the soft-lockup detector generate 5946 backtraces on all cpus. 5947 Format: 0 | 1 5948 5949 sonypi.*= [HW] Sony Programmable I/O Control Device driver 5950 See Documentation/admin-guide/laptops/sonypi.rst 5951 5952 spectre_bhi= [X86] Control mitigation of Branch History Injection 5953 (BHI) vulnerability. This setting affects the 5954 deployment of the HW BHI control and the SW BHB 5955 clearing sequence. 5956 5957 on - (default) Enable the HW or SW mitigation 5958 as needed. 5959 off - Disable the mitigation. 5960 5961 spectre_v2= [X86] Control mitigation of Spectre variant 2 5962 (indirect branch speculation) vulnerability. 5963 The default operation protects the kernel from 5964 user space attacks. 5965 5966 on - unconditionally enable, implies 5967 spectre_v2_user=on 5968 off - unconditionally disable, implies 5969 spectre_v2_user=off 5970 auto - kernel detects whether your CPU model is 5971 vulnerable 5972 5973 Selecting 'on' will, and 'auto' may, choose a 5974 mitigation method at run time according to the 5975 CPU, the available microcode, the setting of the 5976 CONFIG_RETPOLINE configuration option, and the 5977 compiler with which the kernel was built. 5978 5979 Selecting 'on' will also enable the mitigation 5980 against user space to user space task attacks. 5981 5982 Selecting 'off' will disable both the kernel and 5983 the user space protections. 5984 5985 Specific mitigations can also be selected manually: 5986 5987 retpoline - replace indirect branches 5988 retpoline,generic - Retpolines 5989 retpoline,lfence - LFENCE; indirect branch 5990 retpoline,amd - alias for retpoline,lfence 5991 eibrs - Enhanced/Auto IBRS 5992 eibrs,retpoline - Enhanced/Auto IBRS + Retpolines 5993 eibrs,lfence - Enhanced/Auto IBRS + LFENCE 5994 ibrs - use IBRS to protect kernel 5995 5996 Not specifying this option is equivalent to 5997 spectre_v2=auto. 5998 5999 spectre_v2_user= 6000 [X86] Control mitigation of Spectre variant 2 6001 (indirect branch speculation) vulnerability between 6002 user space tasks 6003 6004 on - Unconditionally enable mitigations. Is 6005 enforced by spectre_v2=on 6006 6007 off - Unconditionally disable mitigations. Is 6008 enforced by spectre_v2=off 6009 6010 prctl - Indirect branch speculation is enabled, 6011 but mitigation can be enabled via prctl 6012 per thread. The mitigation control state 6013 is inherited on fork. 6014 6015 prctl,ibpb 6016 - Like "prctl" above, but only STIBP is 6017 controlled per thread. IBPB is issued 6018 always when switching between different user 6019 space processes. 6020 6021 seccomp 6022 - Same as "prctl" above, but all seccomp 6023 threads will enable the mitigation unless 6024 they explicitly opt out. 6025 6026 seccomp,ibpb 6027 - Like "seccomp" above, but only STIBP is 6028 controlled per thread. IBPB is issued 6029 always when switching between different 6030 user space processes. 6031 6032 auto - Kernel selects the mitigation depending on 6033 the available CPU features and vulnerability. 6034 6035 Default mitigation: "prctl" 6036 6037 Not specifying this option is equivalent to 6038 spectre_v2_user=auto. 6039 6040 spec_rstack_overflow= 6041 [X86] Control RAS overflow mitigation on AMD Zen CPUs 6042 6043 off - Disable mitigation 6044 microcode - Enable microcode mitigation only 6045 safe-ret - Enable sw-only safe RET mitigation (default) 6046 ibpb - Enable mitigation by issuing IBPB on 6047 kernel entry 6048 ibpb-vmexit - Issue IBPB only on VMEXIT 6049 (cloud-specific mitigation) 6050 6051 spec_store_bypass_disable= 6052 [HW] Control Speculative Store Bypass (SSB) Disable mitigation 6053 (Speculative Store Bypass vulnerability) 6054 6055 Certain CPUs are vulnerable to an exploit against a 6056 a common industry wide performance optimization known 6057 as "Speculative Store Bypass" in which recent stores 6058 to the same memory location may not be observed by 6059 later loads during speculative execution. The idea 6060 is that such stores are unlikely and that they can 6061 be detected prior to instruction retirement at the 6062 end of a particular speculation execution window. 6063 6064 In vulnerable processors, the speculatively forwarded 6065 store can be used in a cache side channel attack, for 6066 example to read memory to which the attacker does not 6067 directly have access (e.g. inside sandboxed code). 6068 6069 This parameter controls whether the Speculative Store 6070 Bypass optimization is used. 6071 6072 On x86 the options are: 6073 6074 on - Unconditionally disable Speculative Store Bypass 6075 off - Unconditionally enable Speculative Store Bypass 6076 auto - Kernel detects whether the CPU model contains an 6077 implementation of Speculative Store Bypass and 6078 picks the most appropriate mitigation. If the 6079 CPU is not vulnerable, "off" is selected. If the 6080 CPU is vulnerable the default mitigation is 6081 architecture and Kconfig dependent. See below. 6082 prctl - Control Speculative Store Bypass per thread 6083 via prctl. Speculative Store Bypass is enabled 6084 for a process by default. The state of the control 6085 is inherited on fork. 6086 seccomp - Same as "prctl" above, but all seccomp threads 6087 will disable SSB unless they explicitly opt out. 6088 6089 Default mitigations: 6090 X86: "prctl" 6091 6092 On powerpc the options are: 6093 6094 on,auto - On Power8 and Power9 insert a store-forwarding 6095 barrier on kernel entry and exit. On Power7 6096 perform a software flush on kernel entry and 6097 exit. 6098 off - No action. 6099 6100 Not specifying this option is equivalent to 6101 spec_store_bypass_disable=auto. 6102 6103 spia_io_base= [HW,MTD] 6104 spia_fio_base= 6105 spia_pedr= 6106 spia_peddr= 6107 6108 split_lock_detect= 6109 [X86] Enable split lock detection or bus lock detection 6110 6111 When enabled (and if hardware support is present), atomic 6112 instructions that access data across cache line 6113 boundaries will result in an alignment check exception 6114 for split lock detection or a debug exception for 6115 bus lock detection. 6116 6117 off - not enabled 6118 6119 warn - the kernel will emit rate-limited warnings 6120 about applications triggering the #AC 6121 exception or the #DB exception. This mode is 6122 the default on CPUs that support split lock 6123 detection or bus lock detection. Default 6124 behavior is by #AC if both features are 6125 enabled in hardware. 6126 6127 fatal - the kernel will send SIGBUS to applications 6128 that trigger the #AC exception or the #DB 6129 exception. Default behavior is by #AC if 6130 both features are enabled in hardware. 6131 6132 ratelimit:N - 6133 Set system wide rate limit to N bus locks 6134 per second for bus lock detection. 6135 0 < N <= 1000. 6136 6137 N/A for split lock detection. 6138 6139 6140 If an #AC exception is hit in the kernel or in 6141 firmware (i.e. not while executing in user mode) 6142 the kernel will oops in either "warn" or "fatal" 6143 mode. 6144 6145 #DB exception for bus lock is triggered only when 6146 CPL > 0. 6147 6148 srbds= [X86,INTEL] 6149 Control the Special Register Buffer Data Sampling 6150 (SRBDS) mitigation. 6151 6152 Certain CPUs are vulnerable to an MDS-like 6153 exploit which can leak bits from the random 6154 number generator. 6155 6156 By default, this issue is mitigated by 6157 microcode. However, the microcode fix can cause 6158 the RDRAND and RDSEED instructions to become 6159 much slower. Among other effects, this will 6160 result in reduced throughput from /dev/urandom. 6161 6162 The microcode mitigation can be disabled with 6163 the following option: 6164 6165 off: Disable mitigation and remove 6166 performance impact to RDRAND and RDSEED 6167 6168 srcutree.big_cpu_lim [KNL] 6169 Specifies the number of CPUs constituting a 6170 large system, such that srcu_struct structures 6171 should immediately allocate an srcu_node array. 6172 This kernel-boot parameter defaults to 128, 6173 but takes effect only when the low-order four 6174 bits of srcutree.convert_to_big is equal to 3 6175 (decide at boot). 6176 6177 srcutree.convert_to_big [KNL] 6178 Specifies under what conditions an SRCU tree 6179 srcu_struct structure will be converted to big 6180 form, that is, with an rcu_node tree: 6181 6182 0: Never. 6183 1: At init_srcu_struct() time. 6184 2: When rcutorture decides to. 6185 3: Decide at boot time (default). 6186 0x1X: Above plus if high contention. 6187 6188 Either way, the srcu_node tree will be sized based 6189 on the actual runtime number of CPUs (nr_cpu_ids) 6190 instead of the compile-time CONFIG_NR_CPUS. 6191 6192 srcutree.counter_wrap_check [KNL] 6193 Specifies how frequently to check for 6194 grace-period sequence counter wrap for the 6195 srcu_data structure's ->srcu_gp_seq_needed field. 6196 The greater the number of bits set in this kernel 6197 parameter, the less frequently counter wrap will 6198 be checked for. Note that the bottom two bits 6199 are ignored. 6200 6201 srcutree.exp_holdoff [KNL] 6202 Specifies how many nanoseconds must elapse 6203 since the end of the last SRCU grace period for 6204 a given srcu_struct until the next normal SRCU 6205 grace period will be considered for automatic 6206 expediting. Set to zero to disable automatic 6207 expediting. 6208 6209 srcutree.srcu_max_nodelay [KNL] 6210 Specifies the number of no-delay instances 6211 per jiffy for which the SRCU grace period 6212 worker thread will be rescheduled with zero 6213 delay. Beyond this limit, worker thread will 6214 be rescheduled with a sleep delay of one jiffy. 6215 6216 srcutree.srcu_max_nodelay_phase [KNL] 6217 Specifies the per-grace-period phase, number of 6218 non-sleeping polls of readers. Beyond this limit, 6219 grace period worker thread will be rescheduled 6220 with a sleep delay of one jiffy, between each 6221 rescan of the readers, for a grace period phase. 6222 6223 srcutree.srcu_retry_check_delay [KNL] 6224 Specifies number of microseconds of non-sleeping 6225 delay between each non-sleeping poll of readers. 6226 6227 srcutree.small_contention_lim [KNL] 6228 Specifies the number of update-side contention 6229 events per jiffy will be tolerated before 6230 initiating a conversion of an srcu_struct 6231 structure to big form. Note that the value of 6232 srcutree.convert_to_big must have the 0x10 bit 6233 set for contention-based conversions to occur. 6234 6235 ssbd= [ARM64,HW] 6236 Speculative Store Bypass Disable control 6237 6238 On CPUs that are vulnerable to the Speculative 6239 Store Bypass vulnerability and offer a 6240 firmware based mitigation, this parameter 6241 indicates how the mitigation should be used: 6242 6243 force-on: Unconditionally enable mitigation for 6244 for both kernel and userspace 6245 force-off: Unconditionally disable mitigation for 6246 for both kernel and userspace 6247 kernel: Always enable mitigation in the 6248 kernel, and offer a prctl interface 6249 to allow userspace to register its 6250 interest in being mitigated too. 6251 6252 stack_guard_gap= [MM] 6253 override the default stack gap protection. The value 6254 is in page units and it defines how many pages prior 6255 to (for stacks growing down) resp. after (for stacks 6256 growing up) the main stack are reserved for no other 6257 mapping. Default value is 256 pages. 6258 6259 stack_depot_disable= [KNL] 6260 Setting this to true through kernel command line will 6261 disable the stack depot thereby saving the static memory 6262 consumed by the stack hash table. By default this is set 6263 to false. 6264 6265 stacktrace [FTRACE] 6266 Enabled the stack tracer on boot up. 6267 6268 stacktrace_filter=[function-list] 6269 [FTRACE] Limit the functions that the stack tracer 6270 will trace at boot up. function-list is a comma-separated 6271 list of functions. This list can be changed at run 6272 time by the stack_trace_filter file in the debugfs 6273 tracing directory. Note, this enables stack tracing 6274 and the stacktrace above is not needed. 6275 6276 sti= [PARISC,HW] 6277 Format: <num> 6278 Set the STI (builtin display/keyboard on the HP-PARISC 6279 machines) console (graphic card) which should be used 6280 as the initial boot-console. 6281 See also comment in drivers/video/console/sticore.c. 6282 6283 sti_font= [HW] 6284 See comment in drivers/video/console/sticore.c. 6285 6286 stifb= [HW] 6287 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]] 6288 6289 strict_sas_size= 6290 [X86] 6291 Format: <bool> 6292 Enable or disable strict sigaltstack size checks 6293 against the required signal frame size which 6294 depends on the supported FPU features. This can 6295 be used to filter out binaries which have 6296 not yet been made aware of AT_MINSIGSTKSZ. 6297 6298 stress_hpt [PPC] 6299 Limits the number of kernel HPT entries in the hash 6300 page table to increase the rate of hash page table 6301 faults on kernel addresses. 6302 6303 stress_slb [PPC] 6304 Limits the number of kernel SLB entries, and flushes 6305 them frequently to increase the rate of SLB faults 6306 on kernel addresses. 6307 6308 sunrpc.min_resvport= 6309 sunrpc.max_resvport= 6310 [NFS,SUNRPC] 6311 SunRPC servers often require that client requests 6312 originate from a privileged port (i.e. a port in the 6313 range 0 < portnr < 1024). 6314 An administrator who wishes to reserve some of these 6315 ports for other uses may adjust the range that the 6316 kernel's sunrpc client considers to be privileged 6317 using these two parameters to set the minimum and 6318 maximum port values. 6319 6320 sunrpc.svc_rpc_per_connection_limit= 6321 [NFS,SUNRPC] 6322 Limit the number of requests that the server will 6323 process in parallel from a single connection. 6324 The default value is 0 (no limit). 6325 6326 sunrpc.pool_mode= 6327 [NFS] 6328 Control how the NFS server code allocates CPUs to 6329 service thread pools. Depending on how many NICs 6330 you have and where their interrupts are bound, this 6331 option will affect which CPUs will do NFS serving. 6332 Note: this parameter cannot be changed while the 6333 NFS server is running. 6334 6335 auto the server chooses an appropriate mode 6336 automatically using heuristics 6337 global a single global pool contains all CPUs 6338 percpu one pool for each CPU 6339 pernode one pool for each NUMA node (equivalent 6340 to global on non-NUMA machines) 6341 6342 sunrpc.tcp_slot_table_entries= 6343 sunrpc.udp_slot_table_entries= 6344 [NFS,SUNRPC] 6345 Sets the upper limit on the number of simultaneous 6346 RPC calls that can be sent from the client to a 6347 server. Increasing these values may allow you to 6348 improve throughput, but will also increase the 6349 amount of memory reserved for use by the client. 6350 6351 suspend.pm_test_delay= 6352 [SUSPEND] 6353 Sets the number of seconds to remain in a suspend test 6354 mode before resuming the system (see 6355 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG 6356 is set. Default value is 5. 6357 6358 svm= [PPC] 6359 Format: { on | off | y | n | 1 | 0 } 6360 This parameter controls use of the Protected 6361 Execution Facility on pSeries. 6362 6363 swiotlb= [ARM,IA-64,PPC,MIPS,X86] 6364 Format: { <int> [,<int>] | force | noforce } 6365 <int> -- Number of I/O TLB slabs 6366 <int> -- Second integer after comma. Number of swiotlb 6367 areas with their own lock. Will be rounded up 6368 to a power of 2. 6369 force -- force using of bounce buffers even if they 6370 wouldn't be automatically used by the kernel 6371 noforce -- Never use bounce buffers (for debugging) 6372 6373 switches= [HW,M68k] 6374 6375 sysctl.*= [KNL] 6376 Set a sysctl parameter, right before loading the init 6377 process, as if the value was written to the respective 6378 /proc/sys/... file. Both '.' and '/' are recognized as 6379 separators. Unrecognized parameters and invalid values 6380 are reported in the kernel log. Sysctls registered 6381 later by a loaded module cannot be set this way. 6382 Example: sysctl.vm.swappiness=40 6383 6384 sysrq_always_enabled 6385 [KNL] 6386 Ignore sysrq setting - this boot parameter will 6387 neutralize any effect of /proc/sys/kernel/sysrq. 6388 Useful for debugging. 6389 6390 tcpmhash_entries= [KNL,NET] 6391 Set the number of tcp_metrics_hash slots. 6392 Default value is 8192 or 16384 depending on total 6393 ram pages. This is used to specify the TCP metrics 6394 cache size. See Documentation/networking/ip-sysctl.rst 6395 "tcp_no_metrics_save" section for more details. 6396 6397 tdfx= [HW,DRM] 6398 6399 test_suspend= [SUSPEND] 6400 Format: { "mem" | "standby" | "freeze" }[,N] 6401 Specify "mem" (for Suspend-to-RAM) or "standby" (for 6402 standby suspend) or "freeze" (for suspend type freeze) 6403 as the system sleep state during system startup with 6404 the optional capability to repeat N number of times. 6405 The system is woken from this state using a 6406 wakeup-capable RTC alarm. 6407 6408 thash_entries= [KNL,NET] 6409 Set number of hash buckets for TCP connection 6410 6411 thermal.act= [HW,ACPI] 6412 -1: disable all active trip points in all thermal zones 6413 <degrees C>: override all lowest active trip points 6414 6415 thermal.crt= [HW,ACPI] 6416 -1: disable all critical trip points in all thermal zones 6417 <degrees C>: override all critical trip points 6418 6419 thermal.off= [HW,ACPI] 6420 1: disable ACPI thermal control 6421 6422 thermal.psv= [HW,ACPI] 6423 -1: disable all passive trip points 6424 <degrees C>: override all passive trip points to this 6425 value 6426 6427 thermal.tzp= [HW,ACPI] 6428 Specify global default ACPI thermal zone polling rate 6429 <deci-seconds>: poll all this frequency 6430 0: no polling (default) 6431 6432 threadirqs [KNL] 6433 Force threading of all interrupt handlers except those 6434 marked explicitly IRQF_NO_THREAD. 6435 6436 topology= [S390] 6437 Format: {off | on} 6438 Specify if the kernel should make use of the cpu 6439 topology information if the hardware supports this. 6440 The scheduler will make use of this information and 6441 e.g. base its process migration decisions on it. 6442 Default is on. 6443 6444 topology_updates= [KNL, PPC, NUMA] 6445 Format: {off} 6446 Specify if the kernel should ignore (off) 6447 topology updates sent by the hypervisor to this 6448 LPAR. 6449 6450 torture.disable_onoff_at_boot= [KNL] 6451 Prevent the CPU-hotplug component of torturing 6452 until after init has spawned. 6453 6454 torture.ftrace_dump_at_shutdown= [KNL] 6455 Dump the ftrace buffer at torture-test shutdown, 6456 even if there were no errors. This can be a 6457 very costly operation when many torture tests 6458 are running concurrently, especially on systems 6459 with rotating-rust storage. 6460 6461 torture.verbose_sleep_frequency= [KNL] 6462 Specifies how many verbose printk()s should be 6463 emitted between each sleep. The default of zero 6464 disables verbose-printk() sleeping. 6465 6466 torture.verbose_sleep_duration= [KNL] 6467 Duration of each verbose-printk() sleep in jiffies. 6468 6469 tp720= [HW,PS2] 6470 6471 tpm_suspend_pcr=[HW,TPM] 6472 Format: integer pcr id 6473 Specify that at suspend time, the tpm driver 6474 should extend the specified pcr with zeros, 6475 as a workaround for some chips which fail to 6476 flush the last written pcr on TPM_SaveState. 6477 This will guarantee that all the other pcrs 6478 are saved. 6479 6480 tpm_tis.interrupts= [HW,TPM] 6481 Enable interrupts for the MMIO based physical layer 6482 for the FIFO interface. By default it is set to false 6483 (0). For more information about TPM hardware interfaces 6484 defined by Trusted Computing Group (TCG) see 6485 https://trustedcomputinggroup.org/resource/pc-client-platform-tpm-profile-ptp-specification/ 6486 6487 tp_printk [FTRACE] 6488 Have the tracepoints sent to printk as well as the 6489 tracing ring buffer. This is useful for early boot up 6490 where the system hangs or reboots and does not give the 6491 option for reading the tracing buffer or performing a 6492 ftrace_dump_on_oops. 6493 6494 To turn off having tracepoints sent to printk, 6495 echo 0 > /proc/sys/kernel/tracepoint_printk 6496 Note, echoing 1 into this file without the 6497 tracepoint_printk kernel cmdline option has no effect. 6498 6499 The tp_printk_stop_on_boot (see below) can also be used 6500 to stop the printing of events to console at 6501 late_initcall_sync. 6502 6503 ** CAUTION ** 6504 6505 Having tracepoints sent to printk() and activating high 6506 frequency tracepoints such as irq or sched, can cause 6507 the system to live lock. 6508 6509 tp_printk_stop_on_boot [FTRACE] 6510 When tp_printk (above) is set, it can cause a lot of noise 6511 on the console. It may be useful to only include the 6512 printing of events during boot up, as user space may 6513 make the system inoperable. 6514 6515 This command line option will stop the printing of events 6516 to console at the late_initcall_sync() time frame. 6517 6518 trace_buf_size=nn[KMG] 6519 [FTRACE] will set tracing buffer size on each cpu. 6520 6521 trace_clock= [FTRACE] Set the clock used for tracing events 6522 at boot up. 6523 local - Use the per CPU time stamp counter 6524 (converted into nanoseconds). Fast, but 6525 depending on the architecture, may not be 6526 in sync between CPUs. 6527 global - Event time stamps are synchronize across 6528 CPUs. May be slower than the local clock, 6529 but better for some race conditions. 6530 counter - Simple counting of events (1, 2, ..) 6531 note, some counts may be skipped due to the 6532 infrastructure grabbing the clock more than 6533 once per event. 6534 uptime - Use jiffies as the time stamp. 6535 perf - Use the same clock that perf uses. 6536 mono - Use ktime_get_mono_fast_ns() for time stamps. 6537 mono_raw - Use ktime_get_raw_fast_ns() for time 6538 stamps. 6539 boot - Use ktime_get_boot_fast_ns() for time stamps. 6540 Architectures may add more clocks. See 6541 Documentation/trace/ftrace.rst for more details. 6542 6543 trace_event=[event-list] 6544 [FTRACE] Set and start specified trace events in order 6545 to facilitate early boot debugging. The event-list is a 6546 comma-separated list of trace events to enable. See 6547 also Documentation/trace/events.rst 6548 6549 trace_instance=[instance-info] 6550 [FTRACE] Create a ring buffer instance early in boot up. 6551 This will be listed in: 6552 6553 /sys/kernel/tracing/instances 6554 6555 Events can be enabled at the time the instance is created 6556 via: 6557 6558 trace_instance=<name>,<system1>:<event1>,<system2>:<event2> 6559 6560 Note, the "<system*>:" portion is optional if the event is 6561 unique. 6562 6563 trace_instance=foo,sched:sched_switch,irq_handler_entry,initcall 6564 6565 will enable the "sched_switch" event (note, the "sched:" is optional, and 6566 the same thing would happen if it was left off). The irq_handler_entry 6567 event, and all events under the "initcall" system. 6568 6569 trace_options=[option-list] 6570 [FTRACE] Enable or disable tracer options at boot. 6571 The option-list is a comma delimited list of options 6572 that can be enabled or disabled just as if you were 6573 to echo the option name into 6574 6575 /sys/kernel/tracing/trace_options 6576 6577 For example, to enable stacktrace option (to dump the 6578 stack trace of each event), add to the command line: 6579 6580 trace_options=stacktrace 6581 6582 See also Documentation/trace/ftrace.rst "trace options" 6583 section. 6584 6585 trace_trigger=[trigger-list] 6586 [FTRACE] Add a event trigger on specific events. 6587 Set a trigger on top of a specific event, with an optional 6588 filter. 6589 6590 The format is is "trace_trigger=<event>.<trigger>[ if <filter>],..." 6591 Where more than one trigger may be specified that are comma deliminated. 6592 6593 For example: 6594 6595 trace_trigger="sched_switch.stacktrace if prev_state == 2" 6596 6597 The above will enable the "stacktrace" trigger on the "sched_switch" 6598 event but only trigger it if the "prev_state" of the "sched_switch" 6599 event is "2" (TASK_UNINTERUPTIBLE). 6600 6601 See also "Event triggers" in Documentation/trace/events.rst 6602 6603 6604 traceoff_on_warning 6605 [FTRACE] enable this option to disable tracing when a 6606 warning is hit. This turns off "tracing_on". Tracing can 6607 be enabled again by echoing '1' into the "tracing_on" 6608 file located in /sys/kernel/tracing/ 6609 6610 This option is useful, as it disables the trace before 6611 the WARNING dump is called, which prevents the trace to 6612 be filled with content caused by the warning output. 6613 6614 This option can also be set at run time via the sysctl 6615 option: kernel/traceoff_on_warning 6616 6617 transparent_hugepage= 6618 [KNL] 6619 Format: [always|madvise|never] 6620 Can be used to control the default behavior of the system 6621 with respect to transparent hugepages. 6622 See Documentation/admin-guide/mm/transhuge.rst 6623 for more details. 6624 6625 trusted.source= [KEYS] 6626 Format: <string> 6627 This parameter identifies the trust source as a backend 6628 for trusted keys implementation. Supported trust 6629 sources: 6630 - "tpm" 6631 - "tee" 6632 - "caam" 6633 If not specified then it defaults to iterating through 6634 the trust source list starting with TPM and assigns the 6635 first trust source as a backend which is initialized 6636 successfully during iteration. 6637 6638 trusted.rng= [KEYS] 6639 Format: <string> 6640 The RNG used to generate key material for trusted keys. 6641 Can be one of: 6642 - "kernel" 6643 - the same value as trusted.source: "tpm" or "tee" 6644 - "default" 6645 If not specified, "default" is used. In this case, 6646 the RNG's choice is left to each individual trust source. 6647 6648 tsa= [X86] Control mitigation for Transient Scheduler 6649 Attacks on AMD CPUs. Search the following in your 6650 favourite search engine for more details: 6651 6652 "Technical guidance for mitigating transient scheduler 6653 attacks". 6654 6655 off - disable the mitigation 6656 on - enable the mitigation (default) 6657 user - mitigate only user/kernel transitions 6658 vm - mitigate only guest/host transitions 6659 6660 6661 tsc= Disable clocksource stability checks for TSC. 6662 Format: <string> 6663 [x86] reliable: mark tsc clocksource as reliable, this 6664 disables clocksource verification at runtime, as well 6665 as the stability checks done at bootup. Used to enable 6666 high-resolution timer mode on older hardware, and in 6667 virtualized environment. 6668 [x86] noirqtime: Do not use TSC to do irq accounting. 6669 Used to run time disable IRQ_TIME_ACCOUNTING on any 6670 platforms where RDTSC is slow and this accounting 6671 can add overhead. 6672 [x86] unstable: mark the TSC clocksource as unstable, this 6673 marks the TSC unconditionally unstable at bootup and 6674 avoids any further wobbles once the TSC watchdog notices. 6675 [x86] nowatchdog: disable clocksource watchdog. Used 6676 in situations with strict latency requirements (where 6677 interruptions from clocksource watchdog are not 6678 acceptable). 6679 [x86] recalibrate: force recalibration against a HW timer 6680 (HPET or PM timer) on systems whose TSC frequency was 6681 obtained from HW or FW using either an MSR or CPUID(0x15). 6682 Warn if the difference is more than 500 ppm. 6683 [x86] watchdog: Use TSC as the watchdog clocksource with 6684 which to check other HW timers (HPET or PM timer), but 6685 only on systems where TSC has been deemed trustworthy. 6686 This will be suppressed by an earlier tsc=nowatchdog and 6687 can be overridden by a later tsc=nowatchdog. A console 6688 message will flag any such suppression or overriding. 6689 6690 tsc_early_khz= [X86] Skip early TSC calibration and use the given 6691 value instead. Useful when the early TSC frequency discovery 6692 procedure is not reliable, such as on overclocked systems 6693 with CPUID.16h support and partial CPUID.15h support. 6694 Format: <unsigned int> 6695 6696 tsx= [X86] Control Transactional Synchronization 6697 Extensions (TSX) feature in Intel processors that 6698 support TSX control. 6699 6700 This parameter controls the TSX feature. The options are: 6701 6702 on - Enable TSX on the system. Although there are 6703 mitigations for all known security vulnerabilities, 6704 TSX has been known to be an accelerator for 6705 several previous speculation-related CVEs, and 6706 so there may be unknown security risks associated 6707 with leaving it enabled. 6708 6709 off - Disable TSX on the system. (Note that this 6710 option takes effect only on newer CPUs which are 6711 not vulnerable to MDS, i.e., have 6712 MSR_IA32_ARCH_CAPABILITIES.MDS_NO=1 and which get 6713 the new IA32_TSX_CTRL MSR through a microcode 6714 update. This new MSR allows for the reliable 6715 deactivation of the TSX functionality.) 6716 6717 auto - Disable TSX if X86_BUG_TAA is present, 6718 otherwise enable TSX on the system. 6719 6720 Not specifying this option is equivalent to tsx=off. 6721 6722 See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst 6723 for more details. 6724 6725 tsx_async_abort= [X86,INTEL] Control mitigation for the TSX Async 6726 Abort (TAA) vulnerability. 6727 6728 Similar to Micro-architectural Data Sampling (MDS) 6729 certain CPUs that support Transactional 6730 Synchronization Extensions (TSX) are vulnerable to an 6731 exploit against CPU internal buffers which can forward 6732 information to a disclosure gadget under certain 6733 conditions. 6734 6735 In vulnerable processors, the speculatively forwarded 6736 data can be used in a cache side channel attack, to 6737 access data to which the attacker does not have direct 6738 access. 6739 6740 This parameter controls the TAA mitigation. The 6741 options are: 6742 6743 full - Enable TAA mitigation on vulnerable CPUs 6744 if TSX is enabled. 6745 6746 full,nosmt - Enable TAA mitigation and disable SMT on 6747 vulnerable CPUs. If TSX is disabled, SMT 6748 is not disabled because CPU is not 6749 vulnerable to cross-thread TAA attacks. 6750 off - Unconditionally disable TAA mitigation 6751 6752 On MDS-affected machines, tsx_async_abort=off can be 6753 prevented by an active MDS mitigation as both vulnerabilities 6754 are mitigated with the same mechanism so in order to disable 6755 this mitigation, you need to specify mds=off too. 6756 6757 Not specifying this option is equivalent to 6758 tsx_async_abort=full. On CPUs which are MDS affected 6759 and deploy MDS mitigation, TAA mitigation is not 6760 required and doesn't provide any additional 6761 mitigation. 6762 6763 For details see: 6764 Documentation/admin-guide/hw-vuln/tsx_async_abort.rst 6765 6766 turbografx.map[2|3]= [HW,JOY] 6767 TurboGraFX parallel port interface 6768 Format: 6769 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7> 6770 See also Documentation/input/devices/joystick-parport.rst 6771 6772 udbg-immortal [PPC] When debugging early kernel crashes that 6773 happen after console_init() and before a proper 6774 console driver takes over, this boot options might 6775 help "seeing" what's going on. 6776 6777 uhash_entries= [KNL,NET] 6778 Set number of hash buckets for UDP/UDP-Lite connections 6779 6780 uhci-hcd.ignore_oc= 6781 [USB] Ignore overcurrent events (default N). 6782 Some badly-designed motherboards generate lots of 6783 bogus events, for ports that aren't wired to 6784 anything. Set this parameter to avoid log spamming. 6785 Note that genuine overcurrent events won't be 6786 reported either. 6787 6788 unknown_nmi_panic 6789 [X86] Cause panic on unknown NMI. 6790 6791 unwind_debug [X86-64] 6792 Enable unwinder debug output. This can be 6793 useful for debugging certain unwinder error 6794 conditions, including corrupt stacks and 6795 bad/missing unwinder metadata. 6796 6797 usbcore.authorized_default= 6798 [USB] Default USB device authorization: 6799 (default -1 = authorized (same as 1), 6800 0 = not authorized, 1 = authorized, 2 = authorized 6801 if device connected to internal port) 6802 6803 usbcore.autosuspend= 6804 [USB] The autosuspend time delay (in seconds) used 6805 for newly-detected USB devices (default 2). This 6806 is the time required before an idle device will be 6807 autosuspended. Devices for which the delay is set 6808 to a negative value won't be autosuspended at all. 6809 6810 usbcore.usbfs_snoop= 6811 [USB] Set to log all usbfs traffic (default 0 = off). 6812 6813 usbcore.usbfs_snoop_max= 6814 [USB] Maximum number of bytes to snoop in each URB 6815 (default = 65536). 6816 6817 usbcore.blinkenlights= 6818 [USB] Set to cycle leds on hubs (default 0 = off). 6819 6820 usbcore.old_scheme_first= 6821 [USB] Start with the old device initialization 6822 scheme (default 0 = off). 6823 6824 usbcore.usbfs_memory_mb= 6825 [USB] Memory limit (in MB) for buffers allocated by 6826 usbfs (default = 16, 0 = max = 2047). 6827 6828 usbcore.use_both_schemes= 6829 [USB] Try the other device initialization scheme 6830 if the first one fails (default 1 = enabled). 6831 6832 usbcore.initial_descriptor_timeout= 6833 [USB] Specifies timeout for the initial 64-byte 6834 USB_REQ_GET_DESCRIPTOR request in milliseconds 6835 (default 5000 = 5.0 seconds). 6836 6837 usbcore.nousb [USB] Disable the USB subsystem 6838 6839 usbcore.quirks= 6840 [USB] A list of quirk entries to augment the built-in 6841 usb core quirk list. List entries are separated by 6842 commas. Each entry has the form 6843 VendorID:ProductID:Flags. The IDs are 4-digit hex 6844 numbers and Flags is a set of letters. Each letter 6845 will change the built-in quirk; setting it if it is 6846 clear and clearing it if it is set. The letters have 6847 the following meanings: 6848 a = USB_QUIRK_STRING_FETCH_255 (string 6849 descriptors must not be fetched using 6850 a 255-byte read); 6851 b = USB_QUIRK_RESET_RESUME (device can't resume 6852 correctly so reset it instead); 6853 c = USB_QUIRK_NO_SET_INTF (device can't handle 6854 Set-Interface requests); 6855 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't 6856 handle its Configuration or Interface 6857 strings); 6858 e = USB_QUIRK_RESET (device can't be reset 6859 (e.g morph devices), don't use reset); 6860 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has 6861 more interface descriptions than the 6862 bNumInterfaces count, and can't handle 6863 talking to these interfaces); 6864 g = USB_QUIRK_DELAY_INIT (device needs a pause 6865 during initialization, after we read 6866 the device descriptor); 6867 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For 6868 high speed and super speed interrupt 6869 endpoints, the USB 2.0 and USB 3.0 spec 6870 require the interval in microframes (1 6871 microframe = 125 microseconds) to be 6872 calculated as interval = 2 ^ 6873 (bInterval-1). 6874 Devices with this quirk report their 6875 bInterval as the result of this 6876 calculation instead of the exponent 6877 variable used in the calculation); 6878 i = USB_QUIRK_DEVICE_QUALIFIER (device can't 6879 handle device_qualifier descriptor 6880 requests); 6881 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device 6882 generates spurious wakeup, ignore 6883 remote wakeup capability); 6884 k = USB_QUIRK_NO_LPM (device can't handle Link 6885 Power Management); 6886 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL 6887 (Device reports its bInterval as linear 6888 frames instead of the USB 2.0 6889 calculation); 6890 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs 6891 to be disconnected before suspend to 6892 prevent spurious wakeup); 6893 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a 6894 pause after every control message); 6895 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra 6896 delay after resetting its port); 6897 p = USB_QUIRK_SHORT_SET_ADDRESS_REQ_TIMEOUT 6898 (Reduce timeout of the SET_ADDRESS 6899 request from 5000 ms to 500 ms); 6900 Example: quirks=0781:5580:bk,0a5c:5834:gij 6901 6902 usbhid.mousepoll= 6903 [USBHID] The interval which mice are to be polled at. 6904 6905 usbhid.jspoll= 6906 [USBHID] The interval which joysticks are to be polled at. 6907 6908 usbhid.kbpoll= 6909 [USBHID] The interval which keyboards are to be polled at. 6910 6911 usb-storage.delay_use= 6912 [UMS] The delay in seconds before a new device is 6913 scanned for Logical Units (default 1). 6914 6915 usb-storage.quirks= 6916 [UMS] A list of quirks entries to supplement or 6917 override the built-in unusual_devs list. List 6918 entries are separated by commas. Each entry has 6919 the form VID:PID:Flags where VID and PID are Vendor 6920 and Product ID values (4-digit hex numbers) and 6921 Flags is a set of characters, each corresponding 6922 to a common usb-storage quirk flag as follows: 6923 a = SANE_SENSE (collect more than 18 bytes 6924 of sense data, not on uas); 6925 b = BAD_SENSE (don't collect more than 18 6926 bytes of sense data, not on uas); 6927 c = FIX_CAPACITY (decrease the reported 6928 device capacity by one sector); 6929 d = NO_READ_DISC_INFO (don't use 6930 READ_DISC_INFO command, not on uas); 6931 e = NO_READ_CAPACITY_16 (don't use 6932 READ_CAPACITY_16 command); 6933 f = NO_REPORT_OPCODES (don't use report opcodes 6934 command, uas only); 6935 g = MAX_SECTORS_240 (don't transfer more than 6936 240 sectors at a time, uas only); 6937 h = CAPACITY_HEURISTICS (decrease the 6938 reported device capacity by one 6939 sector if the number is odd); 6940 i = IGNORE_DEVICE (don't bind to this 6941 device); 6942 j = NO_REPORT_LUNS (don't use report luns 6943 command, uas only); 6944 k = NO_SAME (do not use WRITE_SAME, uas only) 6945 l = NOT_LOCKABLE (don't try to lock and 6946 unlock ejectable media, not on uas); 6947 m = MAX_SECTORS_64 (don't transfer more 6948 than 64 sectors = 32 KB at a time, 6949 not on uas); 6950 n = INITIAL_READ10 (force a retry of the 6951 initial READ(10) command, not on uas); 6952 o = CAPACITY_OK (accept the capacity 6953 reported by the device, not on uas); 6954 p = WRITE_CACHE (the device cache is ON 6955 by default, not on uas); 6956 r = IGNORE_RESIDUE (the device reports 6957 bogus residue values, not on uas); 6958 s = SINGLE_LUN (the device has only one 6959 Logical Unit); 6960 t = NO_ATA_1X (don't allow ATA(12) and ATA(16) 6961 commands, uas only); 6962 u = IGNORE_UAS (don't bind to the uas driver); 6963 w = NO_WP_DETECT (don't test whether the 6964 medium is write-protected). 6965 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE 6966 even if the device claims no cache, 6967 not on uas) 6968 Example: quirks=0419:aaf5:rl,0421:0433:rc 6969 6970 user_debug= [KNL,ARM] 6971 Format: <int> 6972 See arch/arm/Kconfig.debug help text. 6973 1 - undefined instruction events 6974 2 - system calls 6975 4 - invalid data aborts 6976 8 - SIGSEGV faults 6977 16 - SIGBUS faults 6978 Example: user_debug=31 6979 6980 userpte= 6981 [X86] Flags controlling user PTE allocations. 6982 6983 nohigh = do not allocate PTE pages in 6984 HIGHMEM regardless of setting 6985 of CONFIG_HIGHPTE. 6986 6987 vdso= [X86,SH,SPARC] 6988 On X86_32, this is an alias for vdso32=. Otherwise: 6989 6990 vdso=1: enable VDSO (the default) 6991 vdso=0: disable VDSO mapping 6992 6993 vdso32= [X86] Control the 32-bit vDSO 6994 vdso32=1: enable 32-bit VDSO 6995 vdso32=0 or vdso32=2: disable 32-bit VDSO 6996 6997 See the help text for CONFIG_COMPAT_VDSO for more 6998 details. If CONFIG_COMPAT_VDSO is set, the default is 6999 vdso32=0; otherwise, the default is vdso32=1. 7000 7001 For compatibility with older kernels, vdso32=2 is an 7002 alias for vdso32=0. 7003 7004 Try vdso32=0 if you encounter an error that says: 7005 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed! 7006 7007 vector= [IA-64,SMP] 7008 vector=percpu: enable percpu vector domain 7009 7010 video= [FB] Frame buffer configuration 7011 See Documentation/fb/modedb.rst. 7012 7013 video.brightness_switch_enabled= [ACPI] 7014 Format: [0|1] 7015 If set to 1, on receiving an ACPI notify event 7016 generated by hotkey, video driver will adjust brightness 7017 level and then send out the event to user space through 7018 the allocated input device. If set to 0, video driver 7019 will only send out the event without touching backlight 7020 brightness level. 7021 default: 1 7022 7023 virtio_mmio.device= 7024 [VMMIO] Memory mapped virtio (platform) device. 7025 7026 <size>@<baseaddr>:<irq>[:<id>] 7027 where: 7028 <size> := size (can use standard suffixes 7029 like K, M and G) 7030 <baseaddr> := physical base address 7031 <irq> := interrupt number (as passed to 7032 request_irq()) 7033 <id> := (optional) platform device id 7034 example: 7035 virtio_mmio.device=1K@0x100b0000:48:7 7036 7037 Can be used multiple times for multiple devices. 7038 7039 vga= [BOOT,X86-32] Select a particular video mode 7040 See Documentation/arch/x86/boot.rst and 7041 Documentation/admin-guide/svga.rst. 7042 Use vga=ask for menu. 7043 This is actually a boot loader parameter; the value is 7044 passed to the kernel using a special protocol. 7045 7046 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y. 7047 May slow down system boot speed, especially when 7048 enabled on systems with a large amount of memory. 7049 All options are enabled by default, and this 7050 interface is meant to allow for selectively 7051 enabling or disabling specific virtual memory 7052 debugging features. 7053 7054 Available options are: 7055 P Enable page structure init time poisoning 7056 - Disable all of the above options 7057 7058 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact 7059 size of <nn>. This can be used to increase the 7060 minimum size (128MB on x86). It can also be used to 7061 decrease the size and leave more room for directly 7062 mapped kernel RAM. 7063 7064 vmcp_cma=nn[MG] [KNL,S390] 7065 Sets the memory size reserved for contiguous memory 7066 allocations for the vmcp device driver. 7067 7068 vmhalt= [KNL,S390] Perform z/VM CP command after system halt. 7069 Format: <command> 7070 7071 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic. 7072 Format: <command> 7073 7074 vmpoff= [KNL,S390] Perform z/VM CP command after power off. 7075 Format: <command> 7076 7077 vsyscall= [X86-64] 7078 Controls the behavior of vsyscalls (i.e. calls to 7079 fixed addresses of 0xffffffffff600x00 from legacy 7080 code). Most statically-linked binaries and older 7081 versions of glibc use these calls. Because these 7082 functions are at fixed addresses, they make nice 7083 targets for exploits that can control RIP. 7084 7085 emulate Vsyscalls turn into traps and are emulated 7086 reasonably safely. The vsyscall page is 7087 readable. 7088 7089 xonly [default] Vsyscalls turn into traps and are 7090 emulated reasonably safely. The vsyscall 7091 page is not readable. 7092 7093 none Vsyscalls don't work at all. This makes 7094 them quite hard to use for exploits but 7095 might break your system. 7096 7097 vt.color= [VT] Default text color. 7098 Format: 0xYX, X = foreground, Y = background. 7099 Default: 0x07 = light gray on black. 7100 7101 vt.cur_default= [VT] Default cursor shape. 7102 Format: 0xCCBBAA, where AA, BB, and CC are the same as 7103 the parameters of the <Esc>[?A;B;Cc escape sequence; 7104 see VGA-softcursor.txt. Default: 2 = underline. 7105 7106 vt.default_blu= [VT] 7107 Format: <blue0>,<blue1>,<blue2>,...,<blue15> 7108 Change the default blue palette of the console. 7109 This is a 16-member array composed of values 7110 ranging from 0-255. 7111 7112 vt.default_grn= [VT] 7113 Format: <green0>,<green1>,<green2>,...,<green15> 7114 Change the default green palette of the console. 7115 This is a 16-member array composed of values 7116 ranging from 0-255. 7117 7118 vt.default_red= [VT] 7119 Format: <red0>,<red1>,<red2>,...,<red15> 7120 Change the default red palette of the console. 7121 This is a 16-member array composed of values 7122 ranging from 0-255. 7123 7124 vt.default_utf8= 7125 [VT] 7126 Format=<0|1> 7127 Set system-wide default UTF-8 mode for all tty's. 7128 Default is 1, i.e. UTF-8 mode is enabled for all 7129 newly opened terminals. 7130 7131 vt.global_cursor_default= 7132 [VT] 7133 Format=<-1|0|1> 7134 Set system-wide default for whether a cursor 7135 is shown on new VTs. Default is -1, 7136 i.e. cursors will be created by default unless 7137 overridden by individual drivers. 0 will hide 7138 cursors, 1 will display them. 7139 7140 vt.italic= [VT] Default color for italic text; 0-15. 7141 Default: 2 = green. 7142 7143 vt.underline= [VT] Default color for underlined text; 0-15. 7144 Default: 3 = cyan. 7145 7146 watchdog timers [HW,WDT] For information on watchdog timers, 7147 see Documentation/watchdog/watchdog-parameters.rst 7148 or other driver-specific files in the 7149 Documentation/watchdog/ directory. 7150 7151 watchdog_thresh= 7152 [KNL] 7153 Set the hard lockup detector stall duration 7154 threshold in seconds. The soft lockup detector 7155 threshold is set to twice the value. A value of 0 7156 disables both lockup detectors. Default is 10 7157 seconds. 7158 7159 workqueue.unbound_cpus= 7160 [KNL,SMP] Specify to constrain one or some CPUs 7161 to use in unbound workqueues. 7162 Format: <cpu-list> 7163 By default, all online CPUs are available for 7164 unbound workqueues. 7165 7166 workqueue.watchdog_thresh= 7167 If CONFIG_WQ_WATCHDOG is configured, workqueue can 7168 warn stall conditions and dump internal state to 7169 help debugging. 0 disables workqueue stall 7170 detection; otherwise, it's the stall threshold 7171 duration in seconds. The default value is 30 and 7172 it can be updated at runtime by writing to the 7173 corresponding sysfs file. 7174 7175 workqueue.cpu_intensive_thresh_us= 7176 Per-cpu work items which run for longer than this 7177 threshold are automatically considered CPU intensive 7178 and excluded from concurrency management to prevent 7179 them from noticeably delaying other per-cpu work 7180 items. Default is 10000 (10ms). 7181 7182 If CONFIG_WQ_CPU_INTENSIVE_REPORT is set, the kernel 7183 will report the work functions which violate this 7184 threshold repeatedly. They are likely good 7185 candidates for using WQ_UNBOUND workqueues instead. 7186 7187 workqueue.power_efficient 7188 Per-cpu workqueues are generally preferred because 7189 they show better performance thanks to cache 7190 locality; unfortunately, per-cpu workqueues tend to 7191 be more power hungry than unbound workqueues. 7192 7193 Enabling this makes the per-cpu workqueues which 7194 were observed to contribute significantly to power 7195 consumption unbound, leading to measurably lower 7196 power usage at the cost of small performance 7197 overhead. 7198 7199 The default value of this parameter is determined by 7200 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT. 7201 7202 workqueue.default_affinity_scope= 7203 Select the default affinity scope to use for unbound 7204 workqueues. Can be one of "cpu", "smt", "cache", 7205 "numa" and "system". Default is "cache". For more 7206 information, see the Affinity Scopes section in 7207 Documentation/core-api/workqueue.rst. 7208 7209 This can be changed after boot by writing to the 7210 matching /sys/module/workqueue/parameters file. All 7211 workqueues with the "default" affinity scope will be 7212 updated accordignly. 7213 7214 workqueue.debug_force_rr_cpu 7215 Workqueue used to implicitly guarantee that work 7216 items queued without explicit CPU specified are put 7217 on the local CPU. This guarantee is no longer true 7218 and while local CPU is still preferred work items 7219 may be put on foreign CPUs. This debug option 7220 forces round-robin CPU selection to flush out 7221 usages which depend on the now broken guarantee. 7222 When enabled, memory and cache locality will be 7223 impacted. 7224 7225 writecombine= [LOONGARCH] Control the MAT (Memory Access Type) of 7226 ioremap_wc(). 7227 7228 on - Enable writecombine, use WUC for ioremap_wc() 7229 off - Disable writecombine, use SUC for ioremap_wc() 7230 7231 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of 7232 default x2apic cluster mode on platforms 7233 supporting x2apic. 7234 7235 xen_512gb_limit [KNL,X86-64,XEN] 7236 Restricts the kernel running paravirtualized under Xen 7237 to use only up to 512 GB of RAM. The reason to do so is 7238 crash analysis tools and Xen tools for doing domain 7239 save/restore/migration must be enabled to handle larger 7240 domains. 7241 7242 xen_emul_unplug= [HW,X86,XEN] 7243 Unplug Xen emulated devices 7244 Format: [unplug0,][unplug1] 7245 ide-disks -- unplug primary master IDE devices 7246 aux-ide-disks -- unplug non-primary-master IDE devices 7247 nics -- unplug network devices 7248 all -- unplug all emulated devices (NICs and IDE disks) 7249 unnecessary -- unplugging emulated devices is 7250 unnecessary even if the host did not respond to 7251 the unplug protocol 7252 never -- do not unplug even if version check succeeds 7253 7254 xen_legacy_crash [X86,XEN] 7255 Crash from Xen panic notifier, without executing late 7256 panic() code such as dumping handler. 7257 7258 xen_msr_safe= [X86,XEN] 7259 Format: <bool> 7260 Select whether to always use non-faulting (safe) MSR 7261 access functions when running as Xen PV guest. The 7262 default value is controlled by CONFIG_XEN_PV_MSR_SAFE. 7263 7264 xen_nopvspin [X86,XEN] 7265 Disables the qspinlock slowpath using Xen PV optimizations. 7266 This parameter is obsoleted by "nopvspin" parameter, which 7267 has equivalent effect for XEN platform. 7268 7269 xen_nopv [X86] 7270 Disables the PV optimizations forcing the HVM guest to 7271 run as generic HVM guest with no PV drivers. 7272 This option is obsoleted by the "nopv" option, which 7273 has equivalent effect for XEN platform. 7274 7275 xen_no_vector_callback 7276 [KNL,X86,XEN] Disable the vector callback for Xen 7277 event channel interrupts. 7278 7279 xen_scrub_pages= [XEN] 7280 Boolean option to control scrubbing pages before giving them back 7281 to Xen, for use by other domains. Can be also changed at runtime 7282 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages. 7283 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT. 7284 7285 xen_timer_slop= [X86-64,XEN] 7286 Set the timer slop (in nanoseconds) for the virtual Xen 7287 timers (default is 100000). This adjusts the minimum 7288 delta of virtualized Xen timers, where lower values 7289 improve timer resolution at the expense of processing 7290 more timer interrupts. 7291 7292 xen.balloon_boot_timeout= [XEN] 7293 The time (in seconds) to wait before giving up to boot 7294 in case initial ballooning fails to free enough memory. 7295 Applies only when running as HVM or PVH guest and 7296 started with less memory configured than allowed at 7297 max. Default is 180. 7298 7299 xen.event_eoi_delay= [XEN] 7300 How long to delay EOI handling in case of event 7301 storms (jiffies). Default is 10. 7302 7303 xen.event_loop_timeout= [XEN] 7304 After which time (jiffies) the event handling loop 7305 should start to delay EOI handling. Default is 2. 7306 7307 xen.fifo_events= [XEN] 7308 Boolean parameter to disable using fifo event handling 7309 even if available. Normally fifo event handling is 7310 preferred over the 2-level event handling, as it is 7311 fairer and the number of possible event channels is 7312 much higher. Default is on (use fifo events). 7313 7314 xirc2ps_cs= [NET,PCMCIA] 7315 Format: 7316 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]] 7317 7318 xive= [PPC] 7319 By default on POWER9 and above, the kernel will 7320 natively use the XIVE interrupt controller. This option 7321 allows the fallback firmware mode to be used: 7322 7323 off Fallback to firmware control of XIVE interrupt 7324 controller on both pseries and powernv 7325 platforms. Only useful on POWER9 and above. 7326 7327 xive.store-eoi=off [PPC] 7328 By default on POWER10 and above, the kernel will use 7329 stores for EOI handling when the XIVE interrupt mode 7330 is active. This option allows the XIVE driver to use 7331 loads instead, as on POWER9. 7332 7333 xhci-hcd.quirks [USB,KNL] 7334 A hex value specifying bitmask with supplemental xhci 7335 host controller quirks. Meaning of each bit can be 7336 consulted in header drivers/usb/host/xhci.h. 7337 7338 xmon [PPC] 7339 Format: { early | on | rw | ro | off } 7340 Controls if xmon debugger is enabled. Default is off. 7341 Passing only "xmon" is equivalent to "xmon=early". 7342 early Call xmon as early as possible on boot; xmon 7343 debugger is called from setup_arch(). 7344 on xmon debugger hooks will be installed so xmon 7345 is only called on a kernel crash. Default mode, 7346 i.e. either "ro" or "rw" mode, is controlled 7347 with CONFIG_XMON_DEFAULT_RO_MODE. 7348 rw xmon debugger hooks will be installed so xmon 7349 is called only on a kernel crash, mode is write, 7350 meaning SPR registers, memory and, other data 7351 can be written using xmon commands. 7352 ro same as "rw" option above but SPR registers, 7353 memory, and other data can't be written using 7354 xmon commands. 7355 off xmon is disabled. 7356 7357