1 acpi= [HW,ACPI,X86,ARM64] 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] 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, ONLY "acpi=off", "acpi=on" or "acpi=force" 14 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_PCI_COMPONENT 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 PCI/PCI interrupt routing info messages: 64 acpi.debug_layer=0x400000 65 Enable AML "Debug" output, i.e., stores to the Debug 66 object while interpreting AML: 67 acpi.debug_layer=0xffffffff acpi.debug_level=0x2 68 Enable all messages related to ACPI hardware: 69 acpi.debug_layer=0x2 acpi.debug_level=0xffffffff 70 71 Some values produce so much output that the system is 72 unusable. The "log_buf_len" parameter may be useful 73 if you need to capture more output. 74 75 acpi_enforce_resources= [ACPI] 76 { strict | lax | no } 77 Check for resource conflicts between native drivers 78 and ACPI OperationRegions (SystemIO and SystemMemory 79 only). IO ports and memory declared in ACPI might be 80 used by the ACPI subsystem in arbitrary AML code and 81 can interfere with legacy drivers. 82 strict (default): access to resources claimed by ACPI 83 is denied; legacy drivers trying to access reserved 84 resources will fail to bind to device using them. 85 lax: access to resources claimed by ACPI is allowed; 86 legacy drivers trying to access reserved resources 87 will bind successfully but a warning message is logged. 88 no: ACPI OperationRegions are not marked as reserved, 89 no further checks are performed. 90 91 acpi_force_table_verification [HW,ACPI] 92 Enable table checksum verification during early stage. 93 By default, this is disabled due to x86 early mapping 94 size limitation. 95 96 acpi_irq_balance [HW,ACPI] 97 ACPI will balance active IRQs 98 default in APIC mode 99 100 acpi_irq_nobalance [HW,ACPI] 101 ACPI will not move active IRQs (default) 102 default in PIC mode 103 104 acpi_irq_isa= [HW,ACPI] If irq_balance, mark listed IRQs used by ISA 105 Format: <irq>,<irq>... 106 107 acpi_irq_pci= [HW,ACPI] If irq_balance, clear listed IRQs for 108 use by PCI 109 Format: <irq>,<irq>... 110 111 acpi_mask_gpe= [HW,ACPI] 112 Due to the existence of _Lxx/_Exx, some GPEs triggered 113 by unsupported hardware/firmware features can result in 114 GPE floodings that cannot be automatically disabled by 115 the GPE dispatcher. 116 This facility can be used to prevent such uncontrolled 117 GPE floodings. 118 Format: <byte> 119 120 acpi_no_auto_serialize [HW,ACPI] 121 Disable auto-serialization of AML methods 122 AML control methods that contain the opcodes to create 123 named objects will be marked as "Serialized" by the 124 auto-serialization feature. 125 This feature is enabled by default. 126 This option allows to turn off the feature. 127 128 acpi_no_memhotplug [ACPI] Disable memory hotplug. Useful for kdump 129 kernels. 130 131 acpi_no_static_ssdt [HW,ACPI] 132 Disable installation of static SSDTs at early boot time 133 By default, SSDTs contained in the RSDT/XSDT will be 134 installed automatically and they will appear under 135 /sys/firmware/acpi/tables. 136 This option turns off this feature. 137 Note that specifying this option does not affect 138 dynamic table installation which will install SSDT 139 tables to /sys/firmware/acpi/tables/dynamic. 140 141 acpi_no_watchdog [HW,ACPI,WDT] 142 Ignore the ACPI-based watchdog interface (WDAT) and let 143 a native driver control the watchdog device instead. 144 145 acpi_rsdp= [ACPI,EFI,KEXEC] 146 Pass the RSDP address to the kernel, mostly used 147 on machines running EFI runtime service to boot the 148 second kernel for kdump. 149 150 acpi_os_name= [HW,ACPI] Tell ACPI BIOS the name of the OS 151 Format: To spoof as Windows 98: ="Microsoft Windows" 152 153 acpi_rev_override [ACPI] Override the _REV object to return 5 (instead 154 of 2 which is mandated by ACPI 6) as the supported ACPI 155 specification revision (when using this switch, it may 156 be necessary to carry out a cold reboot _twice_ in a 157 row to make it take effect on the platform firmware). 158 159 acpi_osi= [HW,ACPI] Modify list of supported OS interface strings 160 acpi_osi="string1" # add string1 161 acpi_osi="!string2" # remove string2 162 acpi_osi=!* # remove all strings 163 acpi_osi=! # disable all built-in OS vendor 164 strings 165 acpi_osi=!! # enable all built-in OS vendor 166 strings 167 acpi_osi= # disable all strings 168 169 'acpi_osi=!' can be used in combination with single or 170 multiple 'acpi_osi="string1"' to support specific OS 171 vendor string(s). Note that such command can only 172 affect the default state of the OS vendor strings, thus 173 it cannot affect the default state of the feature group 174 strings and the current state of the OS vendor strings, 175 specifying it multiple times through kernel command line 176 is meaningless. This command is useful when one do not 177 care about the state of the feature group strings which 178 should be controlled by the OSPM. 179 Examples: 180 1. 'acpi_osi=! acpi_osi="Windows 2000"' is equivalent 181 to 'acpi_osi="Windows 2000" acpi_osi=!', they all 182 can make '_OSI("Windows 2000")' TRUE. 183 184 'acpi_osi=' cannot be used in combination with other 185 'acpi_osi=' command lines, the _OSI method will not 186 exist in the ACPI namespace. NOTE that such command can 187 only affect the _OSI support state, thus specifying it 188 multiple times through kernel command line is also 189 meaningless. 190 Examples: 191 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)' 192 FALSE. 193 194 'acpi_osi=!*' can be used in combination with single or 195 multiple 'acpi_osi="string1"' to support specific 196 string(s). Note that such command can affect the 197 current state of both the OS vendor strings and the 198 feature group strings, thus specifying it multiple times 199 through kernel command line is meaningful. But it may 200 still not able to affect the final state of a string if 201 there are quirks related to this string. This command 202 is useful when one want to control the state of the 203 feature group strings to debug BIOS issues related to 204 the OSPM features. 205 Examples: 206 1. 'acpi_osi="Module Device" acpi_osi=!*' can make 207 '_OSI("Module Device")' FALSE. 208 2. 'acpi_osi=!* acpi_osi="Module Device"' can make 209 '_OSI("Module Device")' TRUE. 210 3. 'acpi_osi=! acpi_osi=!* acpi_osi="Windows 2000"' is 211 equivalent to 212 'acpi_osi=!* acpi_osi=! acpi_osi="Windows 2000"' 213 and 214 'acpi_osi=!* acpi_osi="Windows 2000" acpi_osi=!', 215 they all will make '_OSI("Windows 2000")' TRUE. 216 217 acpi_pm_good [X86] 218 Override the pmtimer bug detection: force the kernel 219 to assume that this machine's pmtimer latches its value 220 and always returns good values. 221 222 acpi_sci= [HW,ACPI] ACPI System Control Interrupt trigger mode 223 Format: { level | edge | high | low } 224 225 acpi_skip_timer_override [HW,ACPI] 226 Recognize and ignore IRQ0/pin2 Interrupt Override. 227 For broken nForce2 BIOS resulting in XT-PIC timer. 228 229 acpi_sleep= [HW,ACPI] Sleep options 230 Format: { s3_bios, s3_mode, s3_beep, s4_nohwsig, 231 old_ordering, nonvs, sci_force_enable, nobl } 232 See Documentation/power/video.rst for information on 233 s3_bios and s3_mode. 234 s3_beep is for debugging; it makes the PC's speaker beep 235 as soon as the kernel's real-mode entry point is called. 236 s4_nohwsig prevents ACPI hardware signature from being 237 used during resume from hibernation. 238 old_ordering causes the ACPI 1.0 ordering of the _PTS 239 control method, with respect to putting devices into 240 low power states, to be enforced (the ACPI 2.0 ordering 241 of _PTS is used by default). 242 nonvs prevents the kernel from saving/restoring the 243 ACPI NVS memory during suspend/hibernation and resume. 244 sci_force_enable causes the kernel to set SCI_EN directly 245 on resume from S1/S3 (which is against the ACPI spec, 246 but some broken systems don't work without it). 247 nobl causes the internal blacklist of systems known to 248 behave incorrectly in some ways with respect to system 249 suspend and resume to be ignored (use wisely). 250 251 acpi_use_timer_override [HW,ACPI] 252 Use timer override. For some broken Nvidia NF5 boards 253 that require a timer override, but don't have HPET 254 255 add_efi_memmap [EFI; X86] Include EFI memory map in 256 kernel's map of available physical RAM. 257 258 agp= [AGP] 259 { off | try_unsupported } 260 off: disable AGP support 261 try_unsupported: try to drive unsupported chipsets 262 (may crash computer or cause data corruption) 263 264 ALSA [HW,ALSA] 265 See Documentation/sound/alsa-configuration.rst 266 267 alignment= [KNL,ARM] 268 Allow the default userspace alignment fault handler 269 behaviour to be specified. Bit 0 enables warnings, 270 bit 1 enables fixups, and bit 2 sends a segfault. 271 272 align_va_addr= [X86-64] 273 Align virtual addresses by clearing slice [14:12] when 274 allocating a VMA at process creation time. This option 275 gives you up to 3% performance improvement on AMD F15h 276 machines (where it is enabled by default) for a 277 CPU-intensive style benchmark, and it can vary highly in 278 a microbenchmark depending on workload and compiler. 279 280 32: only for 32-bit processes 281 64: only for 64-bit processes 282 on: enable for both 32- and 64-bit processes 283 off: disable for both 32- and 64-bit processes 284 285 alloc_snapshot [FTRACE] 286 Allocate the ftrace snapshot buffer on boot up when the 287 main buffer is allocated. This is handy if debugging 288 and you need to use tracing_snapshot() on boot up, and 289 do not want to use tracing_snapshot_alloc() as it needs 290 to be done where GFP_KERNEL allocations are allowed. 291 292 allow_file_spec_access 293 Allow speculative faults on file backed pages. 294 Speculative faults are enabled only for those vm_ops 295 that implement and return true for allow_speculation 296 callback. 297 298 allow_mismatched_32bit_el0 [ARM64] 299 Allow execve() of 32-bit applications and setting of the 300 PER_LINUX32 personality on systems where only a strict 301 subset of the CPUs support 32-bit EL0. When this 302 parameter is present, the set of CPUs supporting 32-bit 303 EL0 is indicated by /sys/devices/system/cpu/aarch32_el0 304 and hot-unplug operations may be restricted. 305 306 amd_iommu= [HW,X86-64] 307 Pass parameters to the AMD IOMMU driver in the system. 308 Possible values are: 309 fullflush - enable flushing of IO/TLB entries when 310 they are unmapped. Otherwise they are 311 flushed before they will be reused, which 312 is a lot of faster 313 off - do not initialize any AMD IOMMU found in 314 the system 315 force_isolation - Force device isolation for all 316 devices. The IOMMU driver is not 317 allowed anymore to lift isolation 318 requirements as needed. This option 319 does not override iommu=pt 320 321 amd_iommu_dump= [HW,X86-64] 322 Enable AMD IOMMU driver option to dump the ACPI table 323 for AMD IOMMU. With this option enabled, AMD IOMMU 324 driver will print ACPI tables for AMD IOMMU during 325 IOMMU initialization. 326 327 amd_iommu_intr= [HW,X86-64] 328 Specifies one of the following AMD IOMMU interrupt 329 remapping modes: 330 legacy - Use legacy interrupt remapping mode. 331 vapic - Use virtual APIC mode, which allows IOMMU 332 to inject interrupts directly into guest. 333 This mode requires kvm-amd.avic=1. 334 (Default when IOMMU HW support is present.) 335 336 amijoy.map= [HW,JOY] Amiga joystick support 337 Map of devices attached to JOY0DAT and JOY1DAT 338 Format: <a>,<b> 339 See also Documentation/input/joydev/joystick.rst 340 341 analog.map= [HW,JOY] Analog joystick and gamepad support 342 Specifies type or capabilities of an analog joystick 343 connected to one of 16 gameports 344 Format: <type1>,<type2>,..<type16> 345 346 apc= [HW,SPARC] 347 Power management functions (SPARCstation-4/5 + deriv.) 348 Format: noidle 349 Disable APC CPU standby support. SPARCstation-Fox does 350 not play well with APC CPU idle - disable it if you have 351 APC and your system crashes randomly. 352 353 apic= [APIC,X86] Advanced Programmable Interrupt Controller 354 Change the output verbosity while booting 355 Format: { quiet (default) | verbose | debug } 356 Change the amount of debugging information output 357 when initialising the APIC and IO-APIC components. 358 For X86-32, this can also be used to specify an APIC 359 driver name. 360 Format: apic=driver_name 361 Examples: apic=bigsmp 362 363 apic_extnmi= [APIC,X86] External NMI delivery setting 364 Format: { bsp (default) | all | none } 365 bsp: External NMI is delivered only to CPU 0 366 all: External NMIs are broadcast to all CPUs as a 367 backup of CPU 0 368 none: External NMI is masked for all CPUs. This is 369 useful so that a dump capture kernel won't be 370 shot down by NMI 371 372 autoconf= [IPV6] 373 See Documentation/networking/ipv6.rst. 374 375 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller 376 Limit apic dumping. The parameter defines the maximal 377 number of local apics being dumped. Also it is possible 378 to set it to "all" by meaning -- no limit here. 379 Format: { 1 (default) | 2 | ... | all }. 380 The parameter valid if only apic=debug or 381 apic=verbose is specified. 382 Example: apic=debug show_lapic=all 383 384 apm= [APM] Advanced Power Management 385 See header of arch/x86/kernel/apm_32.c. 386 387 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards 388 Format: <io>,<irq>,<nodeID> 389 390 arm64.nobti [ARM64] Unconditionally disable Branch Target 391 Identification support 392 393 arm64.nopauth [ARM64] Unconditionally disable Pointer Authentication 394 support 395 396 arm64.nomte [ARM64] Unconditionally disable Memory Tagging Extension 397 support 398 399 ataflop= [HW,M68k] 400 401 atarimouse= [HW,MOUSE] Atari Mouse 402 403 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess, 404 EzKey and similar keyboards 405 406 atkbd.reset= [HW] Reset keyboard during initialization 407 408 atkbd.set= [HW] Select keyboard code set 409 Format: <int> (2 = AT (default), 3 = PS/2) 410 411 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar 412 keyboards 413 414 atkbd.softraw= [HW] Choose between synthetic and real raw mode 415 Format: <bool> (0 = real, 1 = synthetic (default)) 416 417 atkbd.softrepeat= [HW] 418 Use software keyboard repeat 419 420 audit= [KNL] Enable the audit sub-system 421 Format: { "0" | "1" | "off" | "on" } 422 0 | off - kernel audit is disabled and can not be 423 enabled until the next reboot 424 unset - kernel audit is initialized but disabled and 425 will be fully enabled by the userspace auditd. 426 1 | on - kernel audit is initialized and partially 427 enabled, storing at most audit_backlog_limit 428 messages in RAM until it is fully enabled by the 429 userspace auditd. 430 Default: unset 431 432 audit_backlog_limit= [KNL] Set the audit queue size limit. 433 Format: <int> (must be >=0) 434 Default: 64 435 436 bau= [X86_UV] Enable the BAU on SGI UV. The default 437 behavior is to disable the BAU (i.e. bau=0). 438 Format: { "0" | "1" } 439 0 - Disable the BAU. 440 1 - Enable the BAU. 441 unset - Disable the BAU. 442 443 baycom_epp= [HW,AX25] 444 Format: <io>,<mode> 445 446 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem 447 Format: <io>,<mode> 448 See header of drivers/net/hamradio/baycom_par.c. 449 450 baycom_ser_fdx= [HW,AX25] 451 BayCom Serial Port AX.25 Modem (Full Duplex Mode) 452 Format: <io>,<irq>,<mode>[,<baud>] 453 See header of drivers/net/hamradio/baycom_ser_fdx.c. 454 455 baycom_ser_hdx= [HW,AX25] 456 BayCom Serial Port AX.25 Modem (Half Duplex Mode) 457 Format: <io>,<irq>,<mode> 458 See header of drivers/net/hamradio/baycom_ser_hdx.c. 459 460 blkdevparts= Manual partition parsing of block device(s) for 461 embedded devices based on command line input. 462 See Documentation/block/cmdline-partition.rst 463 464 boot_delay= Milliseconds to delay each printk during boot. 465 Values larger than 10 seconds (10000) are changed to 466 no delay (0). 467 Format: integer 468 469 bootconfig [KNL] 470 Extended command line options can be added to an initrd 471 and this will cause the kernel to look for it. 472 473 See Documentation/admin-guide/bootconfig.rst 474 475 bert_disable [ACPI] 476 Disable BERT OS support on buggy BIOSes. 477 478 bgrt_disable [ACPI][X86] 479 Disable BGRT to avoid flickering OEM logo. 480 481 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards) 482 bttv.radio= Most important insmod options are available as 483 kernel args too. 484 bttv.pll= See Documentation/admin-guide/media/bttv.rst 485 bttv.tuner= 486 487 bulk_remove=off [PPC] This parameter disables the use of the pSeries 488 firmware feature for flushing multiple hpte entries 489 at a time. 490 491 c101= [NET] Moxa C101 synchronous serial card 492 493 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection. 494 Sometimes CPU hardware bugs make them report the cache 495 size incorrectly. The kernel will attempt work arounds 496 to fix known problems, but for some CPUs it is not 497 possible to determine what the correct size should be. 498 This option provides an override for these situations. 499 500 carrier_timeout= 501 [NET] Specifies amount of time (in seconds) that 502 the kernel should wait for a network carrier. By default 503 it waits 120 seconds. 504 505 ca_keys= [KEYS] This parameter identifies a specific key(s) on 506 the system trusted keyring to be used for certificate 507 trust validation. 508 format: { id:<keyid> | builtin } 509 510 cca= [MIPS] Override the kernel pages' cache coherency 511 algorithm. Accepted values range from 0 to 7 512 inclusive. See arch/mips/include/asm/pgtable-bits.h 513 for platform specific values (SB1, Loongson3 and 514 others). 515 516 ccw_timeout_log [S390] 517 See Documentation/s390/common_io.rst for details. 518 519 cgroup_disable= [KNL] Disable a particular controller or optional feature 520 Format: {name of the controller(s) or feature(s) to disable} 521 The effects of cgroup_disable=foo are: 522 - foo isn't auto-mounted if you mount all cgroups in 523 a single hierarchy 524 - foo isn't visible as an individually mountable 525 subsystem 526 - if foo is an optional feature then the feature is 527 disabled and corresponding cgroup files are not 528 created 529 {Currently only "memory" controller deal with this and 530 cut the overhead, others just disable the usage. So 531 only cgroup_disable=memory is actually worthy} 532 Specifying "pressure" disables per-cgroup pressure 533 stall information accounting feature 534 535 cgroup_no_v1= [KNL] Disable cgroup controllers and named hierarchies in v1 536 Format: { { controller | "all" | "named" } 537 [,{ controller | "all" | "named" }...] } 538 Like cgroup_disable, but only applies to cgroup v1; 539 the blacklisted controllers remain available in cgroup2. 540 "all" blacklists all controllers and "named" disables 541 named mounts. Specifying both "all" and "named" disables 542 all v1 hierarchies. 543 544 cgroup.memory= [KNL] Pass options to the cgroup memory controller. 545 Format: <string> 546 nosocket -- Disable socket memory accounting. 547 nokmem -- Disable kernel memory accounting. 548 549 checkreqprot [SELINUX] Set initial checkreqprot flag value. 550 Format: { "0" | "1" } 551 See security/selinux/Kconfig help text. 552 0 -- check protection applied by kernel (includes 553 any implied execute protection). 554 1 -- check protection requested by application. 555 Default value is set via a kernel config option. 556 Value can be changed at runtime via 557 /sys/fs/selinux/checkreqprot. 558 Setting checkreqprot to 1 is deprecated. 559 560 cio_ignore= [S390] 561 See Documentation/s390/common_io.rst for details. 562 clk_ignore_unused 563 [CLK] 564 Prevents the clock framework from automatically gating 565 clocks that have not been explicitly enabled by a Linux 566 device driver but are enabled in hardware at reset or 567 by the bootloader/firmware. Note that this does not 568 force such clocks to be always-on nor does it reserve 569 those clocks in any way. This parameter is useful for 570 debug and development, but should not be needed on a 571 platform with proper driver support. For more 572 information, see Documentation/driver-api/clk.rst. 573 574 clock= [BUGS=X86-32, HW] gettimeofday clocksource override. 575 [Deprecated] 576 Forces specified clocksource (if available) to be used 577 when calculating gettimeofday(). If specified 578 clocksource is not available, it defaults to PIT. 579 Format: { pit | tsc | cyclone | pmtmr } 580 581 clocksource= Override the default clocksource 582 Format: <string> 583 Override the default clocksource and use the clocksource 584 with the name specified. 585 Some clocksource names to choose from, depending on 586 the platform: 587 [all] jiffies (this is the base, fallback clocksource) 588 [ACPI] acpi_pm 589 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2, 590 pxa_timer,timer3,32k_counter,timer0_1 591 [X86-32] pit,hpet,tsc; 592 scx200_hrt on Geode; cyclone on IBM x440 593 [MIPS] MIPS 594 [PARISC] cr16 595 [S390] tod 596 [SH] SuperH 597 [SPARC64] tick 598 [X86-64] hpet,tsc 599 600 clocksource.arm_arch_timer.evtstrm= 601 [ARM,ARM64] 602 Format: <bool> 603 Enable/disable the eventstream feature of the ARM 604 architected timer so that code using WFE-based polling 605 loops can be debugged more effectively on production 606 systems. 607 608 clocksource.max_cswd_read_retries= [KNL] 609 Number of clocksource_watchdog() retries due to 610 external delays before the clock will be marked 611 unstable. Defaults to three retries, that is, 612 four attempts to read the clock under test. 613 614 clearcpuid=BITNUM[,BITNUM...] [X86] 615 Disable CPUID feature X for the kernel. See 616 arch/x86/include/asm/cpufeatures.h for the valid bit 617 numbers. Note the Linux specific bits are not necessarily 618 stable over kernel options, but the vendor specific 619 ones should be. 620 Also note that user programs calling CPUID directly 621 or using the feature without checking anything 622 will still see it. This just prevents it from 623 being used by the kernel or shown in /proc/cpuinfo. 624 Also note the kernel might malfunction if you disable 625 some critical bits. 626 627 cma=nn[MG]@[start[MG][-end[MG]]] 628 [KNL,CMA] 629 Sets the size of kernel global memory area for 630 contiguous memory allocations and optionally the 631 placement constraint by the physical address range of 632 memory allocations. A value of 0 disables CMA 633 altogether. For more information, see 634 kernel/dma/contiguous.c 635 636 cma_pernuma=nn[MG] 637 [ARM64,KNL] 638 Sets the size of kernel per-numa memory area for 639 contiguous memory allocations. A value of 0 disables 640 per-numa CMA altogether. And If this option is not 641 specificed, the default value is 0. 642 With per-numa CMA enabled, DMA users on node nid will 643 first try to allocate buffer from the pernuma area 644 which is located in node nid, if the allocation fails, 645 they will fallback to the global default memory area. 646 647 cmo_free_hint= [PPC] Format: { yes | no } 648 Specify whether pages are marked as being inactive 649 when they are freed. This is used in CMO environments 650 to determine OS memory pressure for page stealing by 651 a hypervisor. 652 Default: yes 653 654 coherent_pool=nn[KMG] [ARM,KNL] 655 Sets the size of memory pool for coherent, atomic dma 656 allocations, by default set to 256K. 657 658 com20020= [HW,NET] ARCnet - COM20020 chipset 659 Format: 660 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]] 661 662 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers) 663 Format: <io>[,<irq>] 664 665 com90xx= [HW,NET] 666 ARCnet - COM90xx chipset (memory-mapped buffers) 667 Format: <io>[,<irq>[,<memstart>]] 668 669 condev= [HW,S390] console device 670 conmode= 671 672 console= [KNL] Output console device and options. 673 674 tty<n> Use the virtual console device <n>. 675 676 ttyS<n>[,options] 677 ttyUSB0[,options] 678 Use the specified serial port. The options are of 679 the form "bbbbpnf", where "bbbb" is the baud rate, 680 "p" is parity ("n", "o", or "e"), "n" is number of 681 bits, and "f" is flow control ("r" for RTS or 682 omit it). Default is "9600n8". 683 684 See Documentation/admin-guide/serial-console.rst for more 685 information. See 686 Documentation/networking/netconsole.rst for an 687 alternative. 688 689 uart[8250],io,<addr>[,options] 690 uart[8250],mmio,<addr>[,options] 691 uart[8250],mmio16,<addr>[,options] 692 uart[8250],mmio32,<addr>[,options] 693 uart[8250],0x<addr>[,options] 694 Start an early, polled-mode console on the 8250/16550 695 UART at the specified I/O port or MMIO address, 696 switching to the matching ttyS device later. 697 MMIO inter-register address stride is either 8-bit 698 (mmio), 16-bit (mmio16), or 32-bit (mmio32). 699 If none of [io|mmio|mmio16|mmio32], <addr> is assumed 700 to be equivalent to 'mmio'. 'options' are specified in 701 the same format described for ttyS above; if unspecified, 702 the h/w is not re-initialized. 703 704 hvc<n> Use the hypervisor console device <n>. This is for 705 both Xen and PowerPC hypervisors. 706 707 If the device connected to the port is not a TTY but a braille 708 device, prepend "brl," before the device type, for instance 709 console=brl,ttyS0 710 For now, only VisioBraille is supported. 711 712 console_msg_format= 713 [KNL] Change console messages format 714 default 715 By default we print messages on consoles in 716 "[time stamp] text\n" format (time stamp may not be 717 printed, depending on CONFIG_PRINTK_TIME or 718 `printk_time' param). 719 syslog 720 Switch to syslog format: "<%u>[time stamp] text\n" 721 IOW, each message will have a facility and loglevel 722 prefix. The format is similar to one used by syslog() 723 syscall, or to executing "dmesg -S --raw" or to reading 724 from /proc/kmsg. 725 726 consoleblank= [KNL] The console blank (screen saver) timeout in 727 seconds. A value of 0 disables the blank timer. 728 Defaults to 0. 729 730 coredump_filter= 731 [KNL] Change the default value for 732 /proc/<pid>/coredump_filter. 733 See also Documentation/filesystems/proc.rst. 734 735 coresight_cpu_debug.enable 736 [ARM,ARM64] 737 Format: <bool> 738 Enable/disable the CPU sampling based debugging. 739 0: default value, disable debugging 740 1: enable debugging at boot time 741 742 cpuidle.off=1 [CPU_IDLE] 743 disable the cpuidle sub-system 744 745 cpuidle.governor= 746 [CPU_IDLE] Name of the cpuidle governor to use. 747 748 cpufreq.off=1 [CPU_FREQ] 749 disable the cpufreq sub-system 750 751 cpufreq.default_governor= 752 [CPU_FREQ] Name of the default cpufreq governor or 753 policy to use. This governor must be registered in the 754 kernel before the cpufreq driver probes. 755 756 cpu_init_udelay=N 757 [X86] Delay for N microsec between assert and de-assert 758 of APIC INIT to start processors. This delay occurs 759 on every CPU online, such as boot, and resume from suspend. 760 Default: 10000 761 762 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver 763 Format: 764 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>] 765 766 crashkernel=size[KMG][@offset[KMG]] 767 [KNL] Using kexec, Linux can switch to a 'crash kernel' 768 upon panic. This parameter reserves the physical 769 memory region [offset, offset + size] for that kernel 770 image. If '@offset' is omitted, then a suitable offset 771 is selected automatically. 772 [KNL, X86-64] Select a region under 4G first, and 773 fall back to reserve region above 4G when '@offset' 774 hasn't been specified. 775 See Documentation/admin-guide/kdump/kdump.rst for further details. 776 777 crashkernel=range1:size1[,range2:size2,...][@offset] 778 [KNL] Same as above, but depends on the memory 779 in the running system. The syntax of range is 780 start-[end] where start and end are both 781 a memory unit (amount[KMG]). See also 782 Documentation/admin-guide/kdump/kdump.rst for an example. 783 784 crashkernel=size[KMG],high 785 [KNL, X86-64] range could be above 4G. Allow kernel 786 to allocate physical memory region from top, so could 787 be above 4G if system have more than 4G ram installed. 788 Otherwise memory region will be allocated below 4G, if 789 available. 790 It will be ignored if crashkernel=X is specified. 791 crashkernel=size[KMG],low 792 [KNL, X86-64] range under 4G. When crashkernel=X,high 793 is passed, kernel could allocate physical memory region 794 above 4G, that cause second kernel crash on system 795 that require some amount of low memory, e.g. swiotlb 796 requires at least 64M+32K low memory, also enough extra 797 low memory is needed to make sure DMA buffers for 32-bit 798 devices won't run out. Kernel would try to allocate at 799 at least 256M below 4G automatically. 800 This one let user to specify own low range under 4G 801 for second kernel instead. 802 0: to disable low allocation. 803 It will be ignored when crashkernel=X,high is not used 804 or memory reserved is below 4G. 805 806 cryptomgr.notests 807 [KNL] Disable crypto self-tests 808 809 cs89x0_dma= [HW,NET] 810 Format: <dma> 811 812 cs89x0_media= [HW,NET] 813 Format: { rj45 | aui | bnc } 814 815 dasd= [HW,NET] 816 See header of drivers/s390/block/dasd_devmap.c. 817 818 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port 819 (one device per port) 820 Format: <port#>,<type> 821 See also Documentation/input/devices/joystick-parport.rst 822 823 ddebug_query= [KNL,DYNAMIC_DEBUG] Enable debug messages at early boot 824 time. See 825 Documentation/admin-guide/dynamic-debug-howto.rst for 826 details. Deprecated, see dyndbg. 827 828 debug [KNL] Enable kernel debugging (events log level). 829 830 debug_boot_weak_hash 831 [KNL] Enable printing [hashed] pointers early in the 832 boot sequence. If enabled, we use a weak hash instead 833 of siphash to hash pointers. Use this option if you are 834 seeing instances of '(___ptrval___)') and need to see a 835 value (hashed pointer) instead. Cryptographically 836 insecure, please do not use on production kernels. 837 838 debug_locks_verbose= 839 [KNL] verbose self-tests 840 Format=<0|1> 841 Print debugging info while doing the locking API 842 self-tests. 843 We default to 0 (no extra messages), setting it to 844 1 will print _a lot_ more information - normally 845 only useful to kernel developers. 846 847 debug_objects [KNL] Enable object debugging 848 849 no_debug_objects 850 [KNL] Disable object debugging 851 852 debug_guardpage_minorder= 853 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this 854 parameter allows control of the order of pages that will 855 be intentionally kept free (and hence protected) by the 856 buddy allocator. Bigger value increase the probability 857 of catching random memory corruption, but reduce the 858 amount of memory for normal system use. The maximum 859 possible value is MAX_ORDER/2. Setting this parameter 860 to 1 or 2 should be enough to identify most random 861 memory corruption problems caused by bugs in kernel or 862 driver code when a CPU writes to (or reads from) a 863 random memory location. Note that there exists a class 864 of memory corruptions problems caused by buggy H/W or 865 F/W or by drivers badly programing DMA (basically when 866 memory is written at bus level and the CPU MMU is 867 bypassed) which are not detectable by 868 CONFIG_DEBUG_PAGEALLOC, hence this option will not help 869 tracking down these problems. 870 871 debug_pagealloc= 872 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this parameter 873 enables the feature at boot time. By default, it is 874 disabled and the system will work mostly the same as a 875 kernel built without CONFIG_DEBUG_PAGEALLOC. 876 Note: to get most of debug_pagealloc error reports, it's 877 useful to also enable the page_owner functionality. 878 on: enable the feature 879 880 debugfs= [KNL] This parameter enables what is exposed to userspace 881 and debugfs internal clients. 882 Format: { on, no-mount, off } 883 on: All functions are enabled. 884 no-mount: 885 Filesystem is not registered but kernel clients can 886 access APIs and a crashkernel can be used to read 887 its content. There is nothing to mount. 888 off: Filesystem is not registered and clients 889 get a -EPERM as result when trying to register files 890 or directories within debugfs. 891 This is equivalent of the runtime functionality if 892 debugfs was not enabled in the kernel at all. 893 Default value is set in build-time with a kernel configuration. 894 895 debugpat [X86] Enable PAT debugging 896 897 default_hugepagesz= 898 [HW] The size of the default HugeTLB page. This is 899 the size represented by the legacy /proc/ hugepages 900 APIs. In addition, this is the default hugetlb size 901 used for shmget(), mmap() and mounting hugetlbfs 902 filesystems. If not specified, defaults to the 903 architecture's default huge page size. Huge page 904 sizes are architecture dependent. See also 905 Documentation/admin-guide/mm/hugetlbpage.rst. 906 Format: size[KMG] 907 908 deferred_probe_timeout= 909 [KNL] Debugging option to set a timeout in seconds for 910 deferred probe to give up waiting on dependencies to 911 probe. Only specific dependencies (subsystems or 912 drivers) that have opted in will be ignored. A timeout of 0 913 will timeout at the end of initcalls. This option will also 914 dump out devices still on the deferred probe list after 915 retrying. 916 917 dfltcc= [HW,S390] 918 Format: { on | off | def_only | inf_only | always } 919 on: s390 zlib hardware support for compression on 920 level 1 and decompression (default) 921 off: No s390 zlib hardware support 922 def_only: s390 zlib hardware support for deflate 923 only (compression on level 1) 924 inf_only: s390 zlib hardware support for inflate 925 only (decompression) 926 always: Same as 'on' but ignores the selected compression 927 level always using hardware support (used for debugging) 928 929 dhash_entries= [KNL] 930 Set number of hash buckets for dentry cache. 931 932 disable_1tb_segments [PPC] 933 Disables the use of 1TB hash page table segments. This 934 causes the kernel to fall back to 256MB segments which 935 can be useful when debugging issues that require an SLB 936 miss to occur. 937 938 disable_dma32= [KNL] 939 Dynamically disable ZONE_DMA32 on kernels compiled with 940 CONFIG_ZONE_DMA32=y. 941 942 stress_slb [PPC] 943 Limits the number of kernel SLB entries, and flushes 944 them frequently to increase the rate of SLB faults 945 on kernel addresses. 946 947 disable= [IPV6] 948 See Documentation/networking/ipv6.rst. 949 950 hardened_usercopy= 951 [KNL] Under CONFIG_HARDENED_USERCOPY, whether 952 hardening is enabled for this boot. Hardened 953 usercopy checking is used to protect the kernel 954 from reading or writing beyond known memory 955 allocation boundaries as a proactive defense 956 against bounds-checking flaws in the kernel's 957 copy_to_user()/copy_from_user() interface. 958 on Perform hardened usercopy checks (default). 959 off Disable hardened usercopy checks. 960 961 disable_radix [PPC] 962 Disable RADIX MMU mode on POWER9 963 964 radix_hcall_invalidate=on [PPC/PSERIES] 965 Disable RADIX GTSE feature and use hcall for TLB 966 invalidate. 967 968 disable_tlbie [PPC] 969 Disable TLBIE instruction. Currently does not work 970 with KVM, with HASH MMU, or with coherent accelerators. 971 972 disable_cpu_apicid= [X86,APIC,SMP] 973 Format: <int> 974 The number of initial APIC ID for the 975 corresponding CPU to be disabled at boot, 976 mostly used for the kdump 2nd kernel to 977 disable BSP to wake up multiple CPUs without 978 causing system reset or hang due to sending 979 INIT from AP to BSP. 980 981 perf_v4_pmi= [X86,INTEL] 982 Format: <bool> 983 Disable Intel PMU counter freezing feature. 984 The feature only exists starting from 985 Arch Perfmon v4 (Skylake and newer). 986 987 disable_ddw [PPC/PSERIES] 988 Disable Dynamic DMA Window support. Use this 989 to workaround buggy firmware. 990 991 disable_ipv6= [IPV6] 992 See Documentation/networking/ipv6.rst. 993 994 disable_mtrr_cleanup [X86] 995 The kernel tries to adjust MTRR layout from continuous 996 to discrete, to make X server driver able to add WB 997 entry later. This parameter disables that. 998 999 disable_mtrr_trim [X86, Intel and AMD only] 1000 By default the kernel will trim any uncacheable 1001 memory out of your available memory pool based on 1002 MTRR settings. This parameter disables that behavior, 1003 possibly causing your machine to run very slowly. 1004 1005 disable_timer_pin_1 [X86] 1006 Disable PIN 1 of APIC timer 1007 Can be useful to work around chipset bugs. 1008 1009 dis_ucode_ldr [X86] Disable the microcode loader. 1010 1011 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support, 1012 this option disables the debugging code at boot. 1013 1014 dma_debug_entries=<number> 1015 This option allows to tune the number of preallocated 1016 entries for DMA-API debugging code. One entry is 1017 required per DMA-API allocation. Use this if the 1018 DMA-API debugging code disables itself because the 1019 architectural default is too low. 1020 1021 dma_debug_driver=<driver_name> 1022 With this option the DMA-API debugging driver 1023 filter feature can be enabled at boot time. Just 1024 pass the driver to filter for as the parameter. 1025 The filter can be disabled or changed to another 1026 driver later using sysfs. 1027 1028 driver_async_probe= [KNL] 1029 List of driver names to be probed asynchronously. 1030 Format: <driver_name1>,<driver_name2>... 1031 1032 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>] 1033 Broken monitors, graphic adapters, KVMs and EDIDless 1034 panels may send no or incorrect EDID data sets. 1035 This parameter allows to specify an EDID data sets 1036 in the /lib/firmware directory that are used instead. 1037 Generic built-in EDID data sets are used, if one of 1038 edid/1024x768.bin, edid/1280x1024.bin, 1039 edid/1680x1050.bin, or edid/1920x1080.bin is given 1040 and no file with the same name exists. Details and 1041 instructions how to build your own EDID data are 1042 available in Documentation/admin-guide/edid.rst. An EDID 1043 data set will only be used for a particular connector, 1044 if its name and a colon are prepended to the EDID 1045 name. Each connector may use a unique EDID data 1046 set by separating the files with a comma. An EDID 1047 data set with no connector name will be used for 1048 any connectors not explicitly specified. 1049 1050 dscc4.setup= [NET] 1051 1052 dt_cpu_ftrs= [PPC] 1053 Format: {"off" | "known"} 1054 Control how the dt_cpu_ftrs device-tree binding is 1055 used for CPU feature discovery and setup (if it 1056 exists). 1057 off: Do not use it, fall back to legacy cpu table. 1058 known: Do not pass through unknown features to guests 1059 or userspace, only those that the kernel is aware of. 1060 1061 dump_apple_properties [X86] 1062 Dump name and content of EFI device properties on 1063 x86 Macs. Useful for driver authors to determine 1064 what data is available or for reverse-engineering. 1065 1066 dyndbg[="val"] [KNL,DYNAMIC_DEBUG] 1067 <module>.dyndbg[="val"] 1068 Enable debug messages at boot time. See 1069 Documentation/admin-guide/dynamic-debug-howto.rst 1070 for details. 1071 1072 nopku [X86] Disable Memory Protection Keys CPU feature found 1073 in some Intel CPUs. 1074 1075 <module>.async_probe [KNL] 1076 Enable asynchronous probe on this module. 1077 1078 early_ioremap_debug [KNL] 1079 Enable debug messages in early_ioremap support. This 1080 is useful for tracking down temporary early mappings 1081 which are not unmapped. 1082 1083 earlycon= [KNL] Output early console device and options. 1084 1085 When used with no options, the early console is 1086 determined by stdout-path property in device tree's 1087 chosen node or the ACPI SPCR table if supported by 1088 the platform. 1089 1090 cdns,<addr>[,options] 1091 Start an early, polled-mode console on a Cadence 1092 (xuartps) serial port at the specified address. Only 1093 supported option is baud rate. If baud rate is not 1094 specified, the serial port must already be setup and 1095 configured. 1096 1097 uart[8250],io,<addr>[,options] 1098 uart[8250],mmio,<addr>[,options] 1099 uart[8250],mmio32,<addr>[,options] 1100 uart[8250],mmio32be,<addr>[,options] 1101 uart[8250],0x<addr>[,options] 1102 Start an early, polled-mode console on the 8250/16550 1103 UART at the specified I/O port or MMIO address. 1104 MMIO inter-register address stride is either 8-bit 1105 (mmio) or 32-bit (mmio32 or mmio32be). 1106 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed 1107 to be equivalent to 'mmio'. 'options' are specified 1108 in the same format described for "console=ttyS<n>"; if 1109 unspecified, the h/w is not initialized. 1110 1111 pl011,<addr> 1112 pl011,mmio32,<addr> 1113 Start an early, polled-mode console on a pl011 serial 1114 port at the specified address. The pl011 serial port 1115 must already be setup and configured. Options are not 1116 yet supported. If 'mmio32' is specified, then only 1117 the driver will use only 32-bit accessors to read/write 1118 the device registers. 1119 1120 meson,<addr> 1121 Start an early, polled-mode console on a meson serial 1122 port at the specified address. The serial port must 1123 already be setup and configured. Options are not yet 1124 supported. 1125 1126 msm_serial,<addr> 1127 Start an early, polled-mode console on an msm serial 1128 port at the specified address. The serial port 1129 must already be setup and configured. Options are not 1130 yet supported. 1131 1132 msm_serial_dm,<addr> 1133 Start an early, polled-mode console on an msm serial 1134 dm port at the specified address. The serial port 1135 must already be setup and configured. Options are not 1136 yet supported. 1137 1138 owl,<addr> 1139 Start an early, polled-mode console on a serial port 1140 of an Actions Semi SoC, such as S500 or S900, at the 1141 specified address. The serial port must already be 1142 setup and configured. Options are not yet supported. 1143 1144 rda,<addr> 1145 Start an early, polled-mode console on a serial port 1146 of an RDA Micro SoC, such as RDA8810PL, at the 1147 specified address. The serial port must already be 1148 setup and configured. Options are not yet supported. 1149 1150 sbi 1151 Use RISC-V SBI (Supervisor Binary Interface) for early 1152 console. 1153 1154 smh Use ARM semihosting calls for early console. 1155 1156 s3c2410,<addr> 1157 s3c2412,<addr> 1158 s3c2440,<addr> 1159 s3c6400,<addr> 1160 s5pv210,<addr> 1161 exynos4210,<addr> 1162 Use early console provided by serial driver available 1163 on Samsung SoCs, requires selecting proper type and 1164 a correct base address of the selected UART port. The 1165 serial port must already be setup and configured. 1166 Options are not yet supported. 1167 1168 lantiq,<addr> 1169 Start an early, polled-mode console on a lantiq serial 1170 (lqasc) port at the specified address. The serial port 1171 must already be setup and configured. Options are not 1172 yet supported. 1173 1174 lpuart,<addr> 1175 lpuart32,<addr> 1176 Use early console provided by Freescale LP UART driver 1177 found on Freescale Vybrid and QorIQ LS1021A processors. 1178 A valid base address must be provided, and the serial 1179 port must already be setup and configured. 1180 1181 ec_imx21,<addr> 1182 ec_imx6q,<addr> 1183 Start an early, polled-mode, output-only console on the 1184 Freescale i.MX UART at the specified address. The UART 1185 must already be setup and configured. 1186 1187 ar3700_uart,<addr> 1188 Start an early, polled-mode console on the 1189 Armada 3700 serial port at the specified 1190 address. The serial port must already be setup 1191 and configured. Options are not yet supported. 1192 1193 qcom_geni,<addr> 1194 Start an early, polled-mode console on a Qualcomm 1195 Generic Interface (GENI) based serial port at the 1196 specified address. The serial port must already be 1197 setup and configured. Options are not yet supported. 1198 1199 efifb,[options] 1200 Start an early, unaccelerated console on the EFI 1201 memory mapped framebuffer (if available). On cache 1202 coherent non-x86 systems that use system memory for 1203 the framebuffer, pass the 'ram' option so that it is 1204 mapped with the correct attributes. 1205 1206 linflex,<addr> 1207 Use early console provided by Freescale LINFlexD UART 1208 serial driver for NXP S32V234 SoCs. A valid base 1209 address must be provided, and the serial port must 1210 already be setup and configured. 1211 1212 earlyprintk= [X86,SH,ARM,M68k,S390] 1213 earlyprintk=vga 1214 earlyprintk=sclp 1215 earlyprintk=xen 1216 earlyprintk=serial[,ttySn[,baudrate]] 1217 earlyprintk=serial[,0x...[,baudrate]] 1218 earlyprintk=ttySn[,baudrate] 1219 earlyprintk=dbgp[debugController#] 1220 earlyprintk=pciserial[,force],bus:device.function[,baudrate] 1221 earlyprintk=xdbc[xhciController#] 1222 1223 earlyprintk is useful when the kernel crashes before 1224 the normal console is initialized. It is not enabled by 1225 default because it has some cosmetic problems. 1226 1227 Append ",keep" to not disable it when the real console 1228 takes over. 1229 1230 Only one of vga, efi, serial, or usb debug port can 1231 be used at a time. 1232 1233 Currently only ttyS0 and ttyS1 may be specified by 1234 name. Other I/O ports may be explicitly specified 1235 on some architectures (x86 and arm at least) by 1236 replacing ttySn with an I/O port address, like this: 1237 earlyprintk=serial,0x1008,115200 1238 You can find the port for a given device in 1239 /proc/tty/driver/serial: 1240 2: uart:ST16650V2 port:00001008 irq:18 ... 1241 1242 Interaction with the standard serial driver is not 1243 very good. 1244 1245 The VGA and EFI output is eventually overwritten by 1246 the real console. 1247 1248 The xen output can only be used by Xen PV guests. 1249 1250 The sclp output can only be used on s390. 1251 1252 The optional "force" to "pciserial" enables use of a 1253 PCI device even when its classcode is not of the 1254 UART class. 1255 1256 edac_report= [HW,EDAC] Control how to report EDAC event 1257 Format: {"on" | "off" | "force"} 1258 on: enable EDAC to report H/W event. May be overridden 1259 by other higher priority error reporting module. 1260 off: disable H/W event reporting through EDAC. 1261 force: enforce the use of EDAC to report H/W event. 1262 default: on. 1263 1264 ekgdboc= [X86,KGDB] Allow early kernel console debugging 1265 ekgdboc=kbd 1266 1267 This is designed to be used in conjunction with 1268 the boot argument: earlyprintk=vga 1269 1270 This parameter works in place of the kgdboc parameter 1271 but can only be used if the backing tty is available 1272 very early in the boot process. For early debugging 1273 via a serial port see kgdboc_earlycon instead. 1274 1275 edd= [EDD] 1276 Format: {"off" | "on" | "skip[mbr]"} 1277 1278 efi= [EFI] 1279 Format: { "debug", "disable_early_pci_dma", 1280 "nochunk", "noruntime", "nosoftreserve", 1281 "novamap", "no_disable_early_pci_dma" } 1282 debug: enable misc debug output. 1283 disable_early_pci_dma: disable the busmaster bit on all 1284 PCI bridges while in the EFI boot stub. 1285 nochunk: disable reading files in "chunks" in the EFI 1286 boot stub, as chunking can cause problems with some 1287 firmware implementations. 1288 noruntime : disable EFI runtime services support 1289 nosoftreserve: The EFI_MEMORY_SP (Specific Purpose) 1290 attribute may cause the kernel to reserve the 1291 memory range for a memory mapping driver to 1292 claim. Specify efi=nosoftreserve to disable this 1293 reservation and treat the memory by its base type 1294 (i.e. EFI_CONVENTIONAL_MEMORY / "System RAM"). 1295 novamap: do not call SetVirtualAddressMap(). 1296 no_disable_early_pci_dma: Leave the busmaster bit set 1297 on all PCI bridges while in the EFI boot stub 1298 1299 efi_no_storage_paranoia [EFI; X86] 1300 Using this parameter you can use more than 50% of 1301 your efi variable storage. Use this parameter only if 1302 you are really sure that your UEFI does sane gc and 1303 fulfills the spec otherwise your board may brick. 1304 1305 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86] 1306 Add arbitrary attribute to specific memory range by 1307 updating original EFI memory map. 1308 Region of memory which aa attribute is added to is 1309 from ss to ss+nn. 1310 1311 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000 1312 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000) 1313 attribute is added to range 0x100000000-0x180000000 and 1314 0x10a0000000-0x1120000000. 1315 1316 If efi_fake_mem=8G@9G:0x40000 is specified, the 1317 EFI_MEMORY_SP(0x40000) attribute is added to 1318 range 0x240000000-0x43fffffff. 1319 1320 Using this parameter you can do debugging of EFI memmap 1321 related features. For example, you can do debugging of 1322 Address Range Mirroring feature even if your box 1323 doesn't support it, or mark specific memory as 1324 "soft reserved". 1325 1326 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT 1327 that is to be dynamically loaded by Linux. If there are 1328 multiple variables with the same name but with different 1329 vendor GUIDs, all of them will be loaded. See 1330 Documentation/admin-guide/acpi/ssdt-overlays.rst for details. 1331 1332 1333 eisa_irq_edge= [PARISC,HW] 1334 See header of drivers/parisc/eisa.c. 1335 1336 elanfreq= [X86-32] 1337 See comment before function elanfreq_setup() in 1338 arch/x86/kernel/cpu/cpufreq/elanfreq.c. 1339 1340 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390] 1341 Specifies physical address of start of kernel core 1342 image elf header and optionally the size. Generally 1343 kexec loader will pass this option to capture kernel. 1344 See Documentation/admin-guide/kdump/kdump.rst for details. 1345 1346 enable_mtrr_cleanup [X86] 1347 The kernel tries to adjust MTRR layout from continuous 1348 to discrete, to make X server driver able to add WB 1349 entry later. This parameter enables that. 1350 1351 enable_timer_pin_1 [X86] 1352 Enable PIN 1 of APIC timer 1353 Can be useful to work around chipset bugs 1354 (in particular on some ATI chipsets). 1355 The kernel tries to set a reasonable default. 1356 1357 enforcing [SELINUX] Set initial enforcing status. 1358 Format: {"0" | "1"} 1359 See security/selinux/Kconfig help text. 1360 0 -- permissive (log only, no denials). 1361 1 -- enforcing (deny and log). 1362 Default value is 0. 1363 Value can be changed at runtime via 1364 /sys/fs/selinux/enforce. 1365 1366 erst_disable [ACPI] 1367 Disable Error Record Serialization Table (ERST) 1368 support. 1369 1370 ether= [HW,NET] Ethernet cards parameters 1371 This option is obsoleted by the "netdev=" option, which 1372 has equivalent usage. See its documentation for details. 1373 1374 evm= [EVM] 1375 Format: { "fix" } 1376 Permit 'security.evm' to be updated regardless of 1377 current integrity status. 1378 1379 failslab= 1380 fail_usercopy= 1381 fail_page_alloc= 1382 fail_make_request=[KNL] 1383 General fault injection mechanism. 1384 Format: <interval>,<probability>,<space>,<times> 1385 See also Documentation/fault-injection/. 1386 1387 fb_tunnels= [NET] 1388 Format: { initns | none } 1389 See Documentation/admin-guide/sysctl/net.rst for 1390 fb_tunnels_only_for_init_ns 1391 1392 floppy= [HW] 1393 See Documentation/admin-guide/blockdev/floppy.rst. 1394 1395 force_pal_cache_flush 1396 [IA-64] Avoid check_sal_cache_flush which may hang on 1397 buggy SAL_CACHE_FLUSH implementations. Using this 1398 parameter will force ia64_sal_cache_flush to call 1399 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH. 1400 1401 forcepae [X86-32] 1402 Forcefully enable Physical Address Extension (PAE). 1403 Many Pentium M systems disable PAE but may have a 1404 functionally usable PAE implementation. 1405 Warning: use of this parameter will taint the kernel 1406 and may cause unknown problems. 1407 1408 ftrace=[tracer] 1409 [FTRACE] will set and start the specified tracer 1410 as early as possible in order to facilitate early 1411 boot debugging. 1412 1413 ftrace_dump_on_oops[=orig_cpu] 1414 [FTRACE] will dump the trace buffers on oops. 1415 If no parameter is passed, ftrace will dump 1416 buffers of all CPUs, but if you pass orig_cpu, it will 1417 dump only the buffer of the CPU that triggered the 1418 oops. 1419 1420 ftrace_filter=[function-list] 1421 [FTRACE] Limit the functions traced by the function 1422 tracer at boot up. function-list is a comma separated 1423 list of functions. This list can be changed at run 1424 time by the set_ftrace_filter file in the debugfs 1425 tracing directory. 1426 1427 ftrace_notrace=[function-list] 1428 [FTRACE] Do not trace the functions specified in 1429 function-list. This list can be changed at run time 1430 by the set_ftrace_notrace file in the debugfs 1431 tracing directory. 1432 1433 ftrace_graph_filter=[function-list] 1434 [FTRACE] Limit the top level callers functions traced 1435 by the function graph tracer at boot up. 1436 function-list is a comma separated list of functions 1437 that can be changed at run time by the 1438 set_graph_function file in the debugfs tracing directory. 1439 1440 ftrace_graph_notrace=[function-list] 1441 [FTRACE] Do not trace from the functions specified in 1442 function-list. This list is a comma separated list of 1443 functions that can be changed at run time by the 1444 set_graph_notrace file in the debugfs tracing directory. 1445 1446 ftrace_graph_max_depth=<uint> 1447 [FTRACE] Used with the function graph tracer. This is 1448 the max depth it will trace into a function. This value 1449 can be changed at run time by the max_graph_depth file 1450 in the tracefs tracing directory. default: 0 (no limit) 1451 1452 fw_devlink= [KNL] Create device links between consumer and supplier 1453 devices by scanning the firmware to infer the 1454 consumer/supplier relationships. This feature is 1455 especially useful when drivers are loaded as modules as 1456 it ensures proper ordering of tasks like device probing 1457 (suppliers first, then consumers), supplier boot state 1458 clean up (only after all consumers have probed), 1459 suspend/resume & runtime PM (consumers first, then 1460 suppliers). 1461 Format: { off | permissive | on | rpm } 1462 off -- Don't create device links from firmware info. 1463 permissive -- Create device links from firmware info 1464 but use it only for ordering boot state clean 1465 up (sync_state() calls). 1466 on -- Create device links from firmware info and use it 1467 to enforce probe and suspend/resume ordering. 1468 rpm -- Like "on", but also use to order runtime PM. 1469 1470 fw_devlink.strict=<bool> 1471 [KNL] Treat all inferred dependencies as mandatory 1472 dependencies. This only applies for fw_devlink=on|rpm. 1473 Format: <bool> 1474 1475 gamecon.map[2|3]= 1476 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad 1477 support via parallel port (up to 5 devices per port) 1478 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5> 1479 See also Documentation/input/devices/joystick-parport.rst 1480 1481 gamma= [HW,DRM] 1482 1483 gart_fix_e820= [X86-64] disable the fix e820 for K8 GART 1484 Format: off | on 1485 default: on 1486 1487 gather_data_sampling= 1488 [X86,INTEL] Control the Gather Data Sampling (GDS) 1489 mitigation. 1490 1491 Gather Data Sampling is a hardware vulnerability which 1492 allows unprivileged speculative access to data which was 1493 previously stored in vector registers. 1494 1495 This issue is mitigated by default in updated microcode. 1496 The mitigation may have a performance impact but can be 1497 disabled. On systems without the microcode mitigation 1498 disabling AVX serves as a mitigation. 1499 1500 force: Disable AVX to mitigate systems without 1501 microcode mitigation. No effect if the microcode 1502 mitigation is present. Known to cause crashes in 1503 userspace with buggy AVX enumeration. 1504 1505 off: Disable GDS mitigation. 1506 1507 gcov_persist= [GCOV] When non-zero (default), profiling data for 1508 kernel modules is saved and remains accessible via 1509 debugfs, even when the module is unloaded/reloaded. 1510 When zero, profiling data is discarded and associated 1511 debugfs files are removed at module unload time. 1512 1513 goldfish [X86] Enable the goldfish android emulator platform. 1514 Don't use this when you are not running on the 1515 android emulator 1516 1517 gpt [EFI] Forces disk with valid GPT signature but 1518 invalid Protective MBR to be treated as GPT. If the 1519 primary GPT is corrupted, it enables the backup/alternate 1520 GPT to be used instead. 1521 1522 grcan.enable0= [HW] Configuration of physical interface 0. Determines 1523 the "Enable 0" bit of the configuration register. 1524 Format: 0 | 1 1525 Default: 0 1526 grcan.enable1= [HW] Configuration of physical interface 1. Determines 1527 the "Enable 0" bit of the configuration register. 1528 Format: 0 | 1 1529 Default: 0 1530 grcan.select= [HW] Select which physical interface to use. 1531 Format: 0 | 1 1532 Default: 0 1533 grcan.txsize= [HW] Sets the size of the tx buffer. 1534 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0. 1535 Default: 1024 1536 grcan.rxsize= [HW] Sets the size of the rx buffer. 1537 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0. 1538 Default: 1024 1539 1540 gpio-mockup.gpio_mockup_ranges 1541 [HW] Sets the ranges of gpiochip of for this device. 1542 Format: <start1>,<end1>,<start2>,<end2>... 1543 1544 hardlockup_all_cpu_backtrace= 1545 [KNL] Should the hard-lockup detector generate 1546 backtraces on all cpus. 1547 Format: 0 | 1 1548 1549 hashdist= [KNL,NUMA] Large hashes allocated during boot 1550 are distributed across NUMA nodes. Defaults on 1551 for 64-bit NUMA, off otherwise. 1552 Format: 0 | 1 (for off | on) 1553 1554 hcl= [IA-64] SGI's Hardware Graph compatibility layer 1555 1556 hd= [EIDE] (E)IDE hard drive subsystem geometry 1557 Format: <cyl>,<head>,<sect> 1558 1559 hest_disable [ACPI] 1560 Disable Hardware Error Source Table (HEST) support; 1561 corresponding firmware-first mode error processing 1562 logic will be disabled. 1563 1564 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact 1565 size of <nn>. This works even on boxes that have no 1566 highmem otherwise. This also works to reduce highmem 1567 size on bigger boxes. 1568 1569 highres= [KNL] Enable/disable high resolution timer mode. 1570 Valid parameters: "on", "off" 1571 Default: "on" 1572 1573 hlt [BUGS=ARM,SH] 1574 1575 hpet= [X86-32,HPET] option to control HPET usage 1576 Format: { enable (default) | disable | force | 1577 verbose } 1578 disable: disable HPET and use PIT instead 1579 force: allow force enabled of undocumented chips (ICH4, 1580 VIA, nVidia) 1581 verbose: show contents of HPET registers during setup 1582 1583 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET 1584 registers. Default set by CONFIG_HPET_MMAP_DEFAULT. 1585 1586 hugetlb_cma= [HW] The size of a cma area used for allocation 1587 of gigantic hugepages. 1588 Format: nn[KMGTPE] 1589 1590 Reserve a cma area of given size and allocate gigantic 1591 hugepages using the cma allocator. If enabled, the 1592 boot-time allocation of gigantic hugepages is skipped. 1593 1594 hugepages= [HW] Number of HugeTLB pages to allocate at boot. 1595 If this follows hugepagesz (below), it specifies 1596 the number of pages of hugepagesz to be allocated. 1597 If this is the first HugeTLB parameter on the command 1598 line, it specifies the number of pages to allocate for 1599 the default huge page size. See also 1600 Documentation/admin-guide/mm/hugetlbpage.rst. 1601 Format: <integer> 1602 1603 hugepagesz= 1604 [HW] The size of the HugeTLB pages. This is used in 1605 conjunction with hugepages (above) to allocate huge 1606 pages of a specific size at boot. The pair 1607 hugepagesz=X hugepages=Y can be specified once for 1608 each supported huge page size. Huge page sizes are 1609 architecture dependent. See also 1610 Documentation/admin-guide/mm/hugetlbpage.rst. 1611 Format: size[KMG] 1612 1613 hung_task_panic= 1614 [KNL] Should the hung task detector generate panics. 1615 Format: 0 | 1 1616 1617 A value of 1 instructs the kernel to panic when a 1618 hung task is detected. The default value is controlled 1619 by the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time 1620 option. The value selected by this boot parameter can 1621 be changed later by the kernel.hung_task_panic sysctl. 1622 1623 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC) 1624 terminal devices. Valid values: 0..8 1625 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs. 1626 If specified, z/VM IUCV HVC accepts connections 1627 from listed z/VM user IDs only. 1628 1629 hvc_dcc.enable= [ARM,ARM64] Enable DCC driver at runtime. For GKI, 1630 disabled at runtime by default to prevent 1631 crashes in devices which do not support DCC. 1632 1633 hv_nopvspin [X86,HYPER_V] Disables the paravirt spinlock optimizations 1634 which allow the hypervisor to 'idle' the 1635 guest on lock contention. 1636 1637 keep_bootcon [KNL] 1638 Do not unregister boot console at start. This is only 1639 useful for debugging when something happens in the window 1640 between unregistering the boot console and initializing 1641 the real console. 1642 1643 i2c_bus= [HW] Override the default board specific I2C bus speed 1644 or register an additional I2C bus that is not 1645 registered from board initialization code. 1646 Format: 1647 <bus_id>,<clkrate> 1648 1649 i8042.debug [HW] Toggle i8042 debug mode 1650 i8042.unmask_kbd_data 1651 [HW] Enable printing of interrupt data from the KBD port 1652 (disabled by default, and as a pre-condition 1653 requires that i8042.debug=1 be enabled) 1654 i8042.direct [HW] Put keyboard port into non-translated mode 1655 i8042.dumbkbd [HW] Pretend that controller can only read data from 1656 keyboard and cannot control its state 1657 (Don't attempt to blink the leds) 1658 i8042.noaux [HW] Don't check for auxiliary (== mouse) port 1659 i8042.nokbd [HW] Don't check/create keyboard port 1660 i8042.noloop [HW] Disable the AUX Loopback command while probing 1661 for the AUX port 1662 i8042.nomux [HW] Don't check presence of an active multiplexing 1663 controller 1664 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX 1665 controllers 1666 i8042.notimeout [HW] Ignore timeout condition signalled by controller 1667 i8042.reset [HW] Reset the controller during init, cleanup and 1668 suspend-to-ram transitions, only during s2r 1669 transitions, or never reset 1670 Format: { 1 | Y | y | 0 | N | n } 1671 1, Y, y: always reset controller 1672 0, N, n: don't ever reset controller 1673 Default: only on s2r transitions on x86; most other 1674 architectures force reset to be always executed 1675 i8042.unlock [HW] Unlock (ignore) the keylock 1676 i8042.kbdreset [HW] Reset device connected to KBD port 1677 i8042.probe_defer 1678 [HW] Allow deferred probing upon i8042 probe errors 1679 1680 i810= [HW,DRM] 1681 1682 i8k.ignore_dmi [HW] Continue probing hardware even if DMI data 1683 indicates that the driver is running on unsupported 1684 hardware. 1685 i8k.force [HW] Activate i8k driver even if SMM BIOS signature 1686 does not match list of supported models. 1687 i8k.power_status 1688 [HW] Report power status in /proc/i8k 1689 (disabled by default) 1690 i8k.restricted [HW] Allow controlling fans only if SYS_ADMIN 1691 capability is set. 1692 1693 i915.invert_brightness= 1694 [DRM] Invert the sense of the variable that is used to 1695 set the brightness of the panel backlight. Normally a 1696 brightness value of 0 indicates backlight switched off, 1697 and the maximum of the brightness value sets the backlight 1698 to maximum brightness. If this parameter is set to 0 1699 (default) and the machine requires it, or this parameter 1700 is set to 1, a brightness value of 0 sets the backlight 1701 to maximum brightness, and the maximum of the brightness 1702 value switches the backlight off. 1703 -1 -- never invert brightness 1704 0 -- machine default 1705 1 -- force brightness inversion 1706 1707 icn= [HW,ISDN] 1708 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]] 1709 1710 ide-core.nodma= [HW] (E)IDE subsystem 1711 Format: =0.0 to prevent dma on hda, =0.1 hdb =1.0 hdc 1712 .vlb_clock .pci_clock .noflush .nohpa .noprobe .nowerr 1713 .cdrom .chs .ignore_cable are additional options 1714 See Documentation/ide/ide.rst. 1715 1716 ide-generic.probe-mask= [HW] (E)IDE subsystem 1717 Format: <int> 1718 Probe mask for legacy ISA IDE ports. Depending on 1719 platform up to 6 ports are supported, enabled by 1720 setting corresponding bits in the mask to 1. The 1721 default value is 0x0, which has a special meaning. 1722 On systems that have PCI, it triggers scanning the 1723 PCI bus for the first and the second port, which 1724 are then probed. On systems without PCI the value 1725 of 0x0 enables probing the two first ports as if it 1726 was 0x3. 1727 1728 ide-pci-generic.all-generic-ide [HW] (E)IDE subsystem 1729 Claim all unknown PCI IDE storage controllers. 1730 1731 idle= [X86] 1732 Format: idle=poll, idle=halt, idle=nomwait 1733 Poll forces a polling idle loop that can slightly 1734 improve the performance of waking up a idle CPU, but 1735 will use a lot of power and make the system run hot. 1736 Not recommended. 1737 idle=halt: Halt is forced to be used for CPU idle. 1738 In such case C2/C3 won't be used again. 1739 idle=nomwait: Disable mwait for CPU C-states 1740 1741 ieee754= [MIPS] Select IEEE Std 754 conformance mode 1742 Format: { strict | legacy | 2008 | relaxed } 1743 Default: strict 1744 1745 Choose which programs will be accepted for execution 1746 based on the IEEE 754 NaN encoding(s) supported by 1747 the FPU and the NaN encoding requested with the value 1748 of an ELF file header flag individually set by each 1749 binary. Hardware implementations are permitted to 1750 support either or both of the legacy and the 2008 NaN 1751 encoding mode. 1752 1753 Available settings are as follows: 1754 strict accept binaries that request a NaN encoding 1755 supported by the FPU 1756 legacy only accept legacy-NaN binaries, if supported 1757 by the FPU 1758 2008 only accept 2008-NaN binaries, if supported 1759 by the FPU 1760 relaxed accept any binaries regardless of whether 1761 supported by the FPU 1762 1763 The FPU emulator is always able to support both NaN 1764 encodings, so if no FPU hardware is present or it has 1765 been disabled with 'nofpu', then the settings of 1766 'legacy' and '2008' strap the emulator accordingly, 1767 'relaxed' straps the emulator for both legacy-NaN and 1768 2008-NaN, whereas 'strict' enables legacy-NaN only on 1769 legacy processors and both NaN encodings on MIPS32 or 1770 MIPS64 CPUs. 1771 1772 The setting for ABS.fmt/NEG.fmt instruction execution 1773 mode generally follows that for the NaN encoding, 1774 except where unsupported by hardware. 1775 1776 ignore_loglevel [KNL] 1777 Ignore loglevel setting - this will print /all/ 1778 kernel messages to the console. Useful for debugging. 1779 We also add it as printk module parameter, so users 1780 could change it dynamically, usually by 1781 /sys/module/printk/parameters/ignore_loglevel. 1782 1783 ignore_rlimit_data 1784 Ignore RLIMIT_DATA setting for data mappings, 1785 print warning at first misuse. Can be changed via 1786 /sys/module/kernel/parameters/ignore_rlimit_data. 1787 1788 ihash_entries= [KNL] 1789 Set number of hash buckets for inode cache. 1790 1791 ima_appraise= [IMA] appraise integrity measurements 1792 Format: { "off" | "enforce" | "fix" | "log" } 1793 default: "enforce" 1794 1795 ima_appraise_tcb [IMA] Deprecated. Use ima_policy= instead. 1796 The builtin appraise policy appraises all files 1797 owned by uid=0. 1798 1799 ima_canonical_fmt [IMA] 1800 Use the canonical format for the binary runtime 1801 measurements, instead of host native format. 1802 1803 ima_hash= [IMA] 1804 Format: { md5 | sha1 | rmd160 | sha256 | sha384 1805 | sha512 | ... } 1806 default: "sha1" 1807 1808 The list of supported hash algorithms is defined 1809 in crypto/hash_info.h. 1810 1811 ima_policy= [IMA] 1812 The builtin policies to load during IMA setup. 1813 Format: "tcb | appraise_tcb | secure_boot | 1814 fail_securely" 1815 1816 The "tcb" policy measures all programs exec'd, files 1817 mmap'd for exec, and all files opened with the read 1818 mode bit set by either the effective uid (euid=0) or 1819 uid=0. 1820 1821 The "appraise_tcb" policy appraises the integrity of 1822 all files owned by root. 1823 1824 The "secure_boot" policy appraises the integrity 1825 of files (eg. kexec kernel image, kernel modules, 1826 firmware, policy, etc) based on file signatures. 1827 1828 The "fail_securely" policy forces file signature 1829 verification failure also on privileged mounted 1830 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE 1831 flag. 1832 1833 ima_tcb [IMA] Deprecated. Use ima_policy= instead. 1834 Load a policy which meets the needs of the Trusted 1835 Computing Base. This means IMA will measure all 1836 programs exec'd, files mmap'd for exec, and all files 1837 opened for read by uid=0. 1838 1839 ima_template= [IMA] 1840 Select one of defined IMA measurements template formats. 1841 Formats: { "ima" | "ima-ng" | "ima-sig" } 1842 Default: "ima-ng" 1843 1844 ima_template_fmt= 1845 [IMA] Define a custom template format. 1846 Format: { "field1|...|fieldN" } 1847 1848 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage 1849 Format: <min_file_size> 1850 Set the minimal file size for using asynchronous hash. 1851 If left unspecified, ahash usage is disabled. 1852 1853 ahash performance varies for different data sizes on 1854 different crypto accelerators. This option can be used 1855 to achieve the best performance for a particular HW. 1856 1857 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size 1858 Format: <bufsize> 1859 Set hashing buffer size. Default: 4k. 1860 1861 ahash performance varies for different chunk sizes on 1862 different crypto accelerators. This option can be used 1863 to achieve best performance for particular HW. 1864 1865 init= [KNL] 1866 Format: <full_path> 1867 Run specified binary instead of /sbin/init as init 1868 process. 1869 1870 initcall_debug [KNL] Trace initcalls as they are executed. Useful 1871 for working out where the kernel is dying during 1872 startup. 1873 1874 initcall_blacklist= [KNL] Do not execute a comma-separated list of 1875 initcall functions. Useful for debugging built-in 1876 modules and initcalls. 1877 1878 initrd= [BOOT] Specify the location of the initial ramdisk 1879 1880 initrdmem= [KNL] Specify a physical address and size from which to 1881 load the initrd. If an initrd is compiled in or 1882 specified in the bootparams, it takes priority over this 1883 setting. 1884 Format: ss[KMG],nn[KMG] 1885 Default is 0, 0 1886 1887 init_on_alloc= [MM] Fill newly allocated pages and heap objects with 1888 zeroes. 1889 Format: 0 | 1 1890 Default set by CONFIG_INIT_ON_ALLOC_DEFAULT_ON. 1891 1892 init_on_free= [MM] Fill freed pages and heap objects with zeroes. 1893 Format: 0 | 1 1894 Default set by CONFIG_INIT_ON_FREE_DEFAULT_ON. 1895 1896 init_pkru= [X86] Specify the default memory protection keys rights 1897 register contents for all processes. 0x55555554 by 1898 default (disallow access to all but pkey 0). Can 1899 override in debugfs after boot. 1900 1901 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver 1902 Format: <irq> 1903 1904 int_pln_enable [X86] Enable power limit notification interrupt 1905 1906 integrity_audit=[IMA] 1907 Format: { "0" | "1" } 1908 0 -- basic integrity auditing messages. (Default) 1909 1 -- additional integrity auditing messages. 1910 1911 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option 1912 on 1913 Enable intel iommu driver. 1914 off 1915 Disable intel iommu driver. 1916 igfx_off [Default Off] 1917 By default, gfx is mapped as normal device. If a gfx 1918 device has a dedicated DMAR unit, the DMAR unit is 1919 bypassed by not enabling DMAR with this option. In 1920 this case, gfx device will use physical address for 1921 DMA. 1922 forcedac [X86-64] 1923 With this option iommu will not optimize to look 1924 for io virtual address below 32-bit forcing dual 1925 address cycle on pci bus for cards supporting greater 1926 than 32-bit addressing. The default is to look 1927 for translation below 32-bit and if not available 1928 then look in the higher range. 1929 strict [Default Off] 1930 With this option on every unmap_single operation will 1931 result in a hardware IOTLB flush operation as opposed 1932 to batching them for performance. 1933 sp_off [Default Off] 1934 By default, super page will be supported if Intel IOMMU 1935 has the capability. With this option, super page will 1936 not be supported. 1937 sm_on [Default Off] 1938 By default, scalable mode will be disabled even if the 1939 hardware advertises that it has support for the scalable 1940 mode translation. With this option set, scalable mode 1941 will be used on hardware which claims to support it. 1942 tboot_noforce [Default Off] 1943 Do not force the Intel IOMMU enabled under tboot. 1944 By default, tboot will force Intel IOMMU on, which 1945 could harm performance of some high-throughput 1946 devices like 40GBit network cards, even if identity 1947 mapping is enabled. 1948 Note that using this option lowers the security 1949 provided by tboot because it makes the system 1950 vulnerable to DMA attacks. 1951 nobounce [Default off] 1952 Disable bounce buffer for untrusted devices such as 1953 the Thunderbolt devices. This will treat the untrusted 1954 devices as the trusted ones, hence might expose security 1955 risks of DMA attacks. 1956 1957 intel_idle.max_cstate= [KNL,HW,ACPI,X86] 1958 0 disables intel_idle and fall back on acpi_idle. 1959 1 to 9 specify maximum depth of C-state. 1960 1961 intel_pstate= [X86] 1962 disable 1963 Do not enable intel_pstate as the default 1964 scaling driver for the supported processors 1965 passive 1966 Use intel_pstate as a scaling driver, but configure it 1967 to work with generic cpufreq governors (instead of 1968 enabling its internal governor). This mode cannot be 1969 used along with the hardware-managed P-states (HWP) 1970 feature. 1971 force 1972 Enable intel_pstate on systems that prohibit it by default 1973 in favor of acpi-cpufreq. Forcing the intel_pstate driver 1974 instead of acpi-cpufreq may disable platform features, such 1975 as thermal controls and power capping, that rely on ACPI 1976 P-States information being indicated to OSPM and therefore 1977 should be used with caution. This option does not work with 1978 processors that aren't supported by the intel_pstate driver 1979 or on platforms that use pcc-cpufreq instead of acpi-cpufreq. 1980 no_hwp 1981 Do not enable hardware P state control (HWP) 1982 if available. 1983 hwp_only 1984 Only load intel_pstate on systems which support 1985 hardware P state control (HWP) if available. 1986 support_acpi_ppc 1987 Enforce ACPI _PPC performance limits. If the Fixed ACPI 1988 Description Table, specifies preferred power management 1989 profile as "Enterprise Server" or "Performance Server", 1990 then this feature is turned on by default. 1991 per_cpu_perf_limits 1992 Allow per-logical-CPU P-State performance control limits using 1993 cpufreq sysfs interface 1994 1995 intremap= [X86-64, Intel-IOMMU] 1996 on enable Interrupt Remapping (default) 1997 off disable Interrupt Remapping 1998 nosid disable Source ID checking 1999 no_x2apic_optout 2000 BIOS x2APIC opt-out request will be ignored 2001 nopost disable Interrupt Posting 2002 2003 iomem= Disable strict checking of access to MMIO memory 2004 strict regions from userspace. 2005 relaxed 2006 2007 iommu= [X86] 2008 off 2009 force 2010 noforce 2011 biomerge 2012 panic 2013 nopanic 2014 merge 2015 nomerge 2016 soft 2017 pt [X86] 2018 nopt [X86] 2019 nobypass [PPC/POWERNV] 2020 Disable IOMMU bypass, using IOMMU for PCI devices. 2021 2022 iommu.strict= [ARM64] Configure TLB invalidation behaviour 2023 Format: { "0" | "1" } 2024 0 - Lazy mode. 2025 Request that DMA unmap operations use deferred 2026 invalidation of hardware TLBs, for increased 2027 throughput at the cost of reduced device isolation. 2028 Will fall back to strict mode if not supported by 2029 the relevant IOMMU driver. 2030 1 - Strict mode (default). 2031 DMA unmap operations invalidate IOMMU hardware TLBs 2032 synchronously. 2033 2034 iommu.passthrough= 2035 [ARM64, X86] Configure DMA to bypass the IOMMU by default. 2036 Format: { "0" | "1" } 2037 0 - Use IOMMU translation for DMA. 2038 1 - Bypass the IOMMU for DMA. 2039 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH. 2040 2041 io7= [HW] IO7 for Marvel-based Alpha systems 2042 See comment before marvel_specify_io7 in 2043 arch/alpha/kernel/core_marvel.c. 2044 2045 io_delay= [X86] I/O delay method 2046 0x80 2047 Standard port 0x80 based delay 2048 0xed 2049 Alternate port 0xed based delay (needed on some systems) 2050 udelay 2051 Simple two microseconds delay 2052 none 2053 No delay 2054 2055 ip= [IP_PNP] 2056 See Documentation/admin-guide/nfs/nfsroot.rst. 2057 2058 ipcmni_extend [KNL] Extend the maximum number of unique System V 2059 IPC identifiers from 32,768 to 16,777,216. 2060 2061 irqaffinity= [SMP] Set the default irq affinity mask 2062 The argument is a cpu list, as described above. 2063 2064 irqchip.gicv2_force_probe= 2065 [ARM, ARM64] 2066 Format: <bool> 2067 Force the kernel to look for the second 4kB page 2068 of a GICv2 controller even if the memory range 2069 exposed by the device tree is too small. 2070 2071 irqchip.gicv3_nolpi= 2072 [ARM, ARM64] 2073 Force the kernel to ignore the availability of 2074 LPIs (and by consequence ITSs). Intended for system 2075 that use the kernel as a bootloader, and thus want 2076 to let secondary kernels in charge of setting up 2077 LPIs. 2078 2079 irqchip.gicv3_pseudo_nmi= [ARM64] 2080 Enables support for pseudo-NMIs in the kernel. This 2081 requires the kernel to be built with 2082 CONFIG_ARM64_PSEUDO_NMI. 2083 2084 irqfixup [HW] 2085 When an interrupt is not handled search all handlers 2086 for it. Intended to get systems with badly broken 2087 firmware running. 2088 2089 irqpoll [HW] 2090 When an interrupt is not handled search all handlers 2091 for it. Also check all handlers each timer 2092 interrupt. Intended to get systems with badly broken 2093 firmware running. 2094 2095 isapnp= [ISAPNP] 2096 Format: <RDP>,<reset>,<pci_scan>,<verbosity> 2097 2098 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance. 2099 [Deprecated - use cpusets instead] 2100 Format: [flag-list,]<cpu-list> 2101 2102 Specify one or more CPUs to isolate from disturbances 2103 specified in the flag list (default: domain): 2104 2105 nohz 2106 Disable the tick when a single task runs. 2107 2108 A residual 1Hz tick is offloaded to workqueues, which you 2109 need to affine to housekeeping through the global 2110 workqueue's affinity configured via the 2111 /sys/devices/virtual/workqueue/cpumask sysfs file, or 2112 by using the 'domain' flag described below. 2113 2114 NOTE: by default the global workqueue runs on all CPUs, 2115 so to protect individual CPUs the 'cpumask' file has to 2116 be configured manually after bootup. 2117 2118 domain 2119 Isolate from the general SMP balancing and scheduling 2120 algorithms. Note that performing domain isolation this way 2121 is irreversible: it's not possible to bring back a CPU to 2122 the domains once isolated through isolcpus. It's strongly 2123 advised to use cpusets instead to disable scheduler load 2124 balancing through the "cpuset.sched_load_balance" file. 2125 It offers a much more flexible interface where CPUs can 2126 move in and out of an isolated set anytime. 2127 2128 You can move a process onto or off an "isolated" CPU via 2129 the CPU affinity syscalls or cpuset. 2130 <cpu number> begins at 0 and the maximum value is 2131 "number of CPUs in system - 1". 2132 2133 managed_irq 2134 2135 Isolate from being targeted by managed interrupts 2136 which have an interrupt mask containing isolated 2137 CPUs. The affinity of managed interrupts is 2138 handled by the kernel and cannot be changed via 2139 the /proc/irq/* interfaces. 2140 2141 This isolation is best effort and only effective 2142 if the automatically assigned interrupt mask of a 2143 device queue contains isolated and housekeeping 2144 CPUs. If housekeeping CPUs are online then such 2145 interrupts are directed to the housekeeping CPU 2146 so that IO submitted on the housekeeping CPU 2147 cannot disturb the isolated CPU. 2148 2149 If a queue's affinity mask contains only isolated 2150 CPUs then this parameter has no effect on the 2151 interrupt routing decision, though interrupts are 2152 only delivered when tasks running on those 2153 isolated CPUs submit IO. IO submitted on 2154 housekeeping CPUs has no influence on those 2155 queues. 2156 2157 The format of <cpu-list> is described above. 2158 2159 iucv= [HW,NET] 2160 2161 ivrs_ioapic [HW,X86-64] 2162 Provide an override to the IOAPIC-ID<->DEVICE-ID 2163 mapping provided in the IVRS ACPI table. 2164 By default, PCI segment is 0, and can be omitted. 2165 2166 For example, to map IOAPIC-ID decimal 10 to 2167 PCI segment 0x1 and PCI device 00:14.0, 2168 write the parameter as: 2169 ivrs_ioapic=10@0001:00:14.0 2170 2171 Deprecated formats: 2172 * To map IOAPIC-ID decimal 10 to PCI device 00:14.0 2173 write the parameter as: 2174 ivrs_ioapic[10]=00:14.0 2175 * To map IOAPIC-ID decimal 10 to PCI segment 0x1 and 2176 PCI device 00:14.0 write the parameter as: 2177 ivrs_ioapic[10]=0001:00:14.0 2178 2179 ivrs_hpet [HW,X86-64] 2180 Provide an override to the HPET-ID<->DEVICE-ID 2181 mapping provided in the IVRS ACPI table. 2182 By default, PCI segment is 0, and can be omitted. 2183 2184 For example, to map HPET-ID decimal 10 to 2185 PCI segment 0x1 and PCI device 00:14.0, 2186 write the parameter as: 2187 ivrs_hpet=10@0001:00:14.0 2188 2189 Deprecated formats: 2190 * To map HPET-ID decimal 0 to PCI device 00:14.0 2191 write the parameter as: 2192 ivrs_hpet[0]=00:14.0 2193 * To map HPET-ID decimal 10 to PCI segment 0x1 and 2194 PCI device 00:14.0 write the parameter as: 2195 ivrs_ioapic[10]=0001:00:14.0 2196 2197 ivrs_acpihid [HW,X86-64] 2198 Provide an override to the ACPI-HID:UID<->DEVICE-ID 2199 mapping provided in the IVRS ACPI table. 2200 By default, PCI segment is 0, and can be omitted. 2201 2202 For example, to map UART-HID:UID AMD0020:0 to 2203 PCI segment 0x1 and PCI device ID 00:14.5, 2204 write the parameter as: 2205 ivrs_acpihid=AMD0020:0@0001:00:14.5 2206 2207 Deprecated formats: 2208 * To map UART-HID:UID AMD0020:0 to PCI segment is 0, 2209 PCI device ID 00:14.5, write the parameter as: 2210 ivrs_acpihid[00:14.5]=AMD0020:0 2211 * To map UART-HID:UID AMD0020:0 to PCI segment 0x1 and 2212 PCI device ID 00:14.5, write the parameter as: 2213 ivrs_acpihid[0001:00:14.5]=AMD0020:0 2214 2215 js= [HW,JOY] Analog joystick 2216 See Documentation/input/joydev/joystick.rst. 2217 2218 nokaslr [KNL] 2219 When CONFIG_RANDOMIZE_BASE is set, this disables 2220 kernel and module base offset ASLR (Address Space 2221 Layout Randomization). 2222 2223 kasan_multi_shot 2224 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print 2225 report on every invalid memory access. Without this 2226 parameter KASAN will print report only for the first 2227 invalid access. 2228 2229 keepinitrd [HW,ARM] 2230 2231 kernelcore= [KNL,X86,IA-64,PPC] 2232 Format: nn[KMGTPE] | nn% | "mirror" 2233 This parameter specifies the amount of memory usable by 2234 the kernel for non-movable allocations. The requested 2235 amount is spread evenly throughout all nodes in the 2236 system as ZONE_NORMAL. The remaining memory is used for 2237 movable memory in its own zone, ZONE_MOVABLE. In the 2238 event, a node is too small to have both ZONE_NORMAL and 2239 ZONE_MOVABLE, kernelcore memory will take priority and 2240 other nodes will have a larger ZONE_MOVABLE. 2241 2242 ZONE_MOVABLE is used for the allocation of pages that 2243 may be reclaimed or moved by the page migration 2244 subsystem. Note that allocations like PTEs-from-HighMem 2245 still use the HighMem zone if it exists, and the Normal 2246 zone if it does not. 2247 2248 It is possible to specify the exact amount of memory in 2249 the form of "nn[KMGTPE]", a percentage of total system 2250 memory in the form of "nn%", or "mirror". If "mirror" 2251 option is specified, mirrored (reliable) memory is used 2252 for non-movable allocations and remaining memory is used 2253 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror" 2254 are exclusive, so you cannot specify multiple forms. 2255 2256 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port. 2257 Format: <Controller#>[,poll interval] 2258 The controller # is the number of the ehci usb debug 2259 port as it is probed via PCI. The poll interval is 2260 optional and is the number seconds in between 2261 each poll cycle to the debug port in case you need 2262 the functionality for interrupting the kernel with 2263 gdb or control-c on the dbgp connection. When 2264 not using this parameter you use sysrq-g to break into 2265 the kernel debugger. 2266 2267 kgdboc= [KGDB,HW] kgdb over consoles. 2268 Requires a tty driver that supports console polling, 2269 or a supported polling keyboard driver (non-usb). 2270 Serial only format: <serial_device>[,baud] 2271 keyboard only format: kbd 2272 keyboard and serial format: kbd,<serial_device>[,baud] 2273 Optional Kernel mode setting: 2274 kms, kbd format: kms,kbd 2275 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud] 2276 2277 kgdboc_earlycon= [KGDB,HW] 2278 If the boot console provides the ability to read 2279 characters and can work in polling mode, you can use 2280 this parameter to tell kgdb to use it as a backend 2281 until the normal console is registered. Intended to 2282 be used together with the kgdboc parameter which 2283 specifies the normal console to transition to. 2284 2285 The name of the early console should be specified 2286 as the value of this parameter. Note that the name of 2287 the early console might be different than the tty 2288 name passed to kgdboc. It's OK to leave the value 2289 blank and the first boot console that implements 2290 read() will be picked. 2291 2292 kgdbwait [KGDB] Stop kernel execution and enter the 2293 kernel debugger at the earliest opportunity. 2294 2295 kmac= [MIPS] Korina ethernet MAC address. 2296 Configure the RouterBoard 532 series on-chip 2297 Ethernet adapter MAC address. 2298 2299 kmemleak= [KNL] Boot-time kmemleak enable/disable 2300 Valid arguments: on, off 2301 Default: on 2302 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y, 2303 the default is off. 2304 2305 kprobe_event=[probe-list] 2306 [FTRACE] Add kprobe events and enable at boot time. 2307 The probe-list is a semicolon delimited list of probe 2308 definitions. Each definition is same as kprobe_events 2309 interface, but the parameters are comma delimited. 2310 For example, to add a kprobe event on vfs_read with 2311 arg1 and arg2, add to the command line; 2312 2313 kprobe_event=p,vfs_read,$arg1,$arg2 2314 2315 See also Documentation/trace/kprobetrace.rst "Kernel 2316 Boot Parameter" section. 2317 2318 kpti= [ARM64] Control page table isolation of user 2319 and kernel address spaces. 2320 Default: enabled on cores which need mitigation. 2321 0: force disabled 2322 1: force enabled 2323 2324 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs. 2325 Default is 0 (don't ignore, but inject #GP) 2326 2327 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface. 2328 Default is false (don't support). 2329 2330 kvm.mmu_audit= [KVM] This is a R/W parameter which allows audit 2331 KVM MMU at runtime. 2332 Default is 0 (off) 2333 2334 kvm.nx_huge_pages= 2335 [KVM] Controls the software workaround for the 2336 X86_BUG_ITLB_MULTIHIT bug. 2337 force : Always deploy workaround. 2338 off : Never deploy workaround. 2339 auto : Deploy workaround based on the presence of 2340 X86_BUG_ITLB_MULTIHIT. 2341 2342 Default is 'auto'. 2343 2344 If the software workaround is enabled for the host, 2345 guests do need not to enable it for nested guests. 2346 2347 kvm.nx_huge_pages_recovery_ratio= 2348 [KVM] Controls how many 4KiB pages are periodically zapped 2349 back to huge pages. 0 disables the recovery, otherwise if 2350 the value is N KVM will zap 1/Nth of the 4KiB pages every 2351 minute. The default is 60. 2352 2353 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM. 2354 Default is 1 (enabled) 2355 2356 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU) 2357 for all guests. 2358 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode. 2359 2360 kvm-arm.mode= 2361 [KVM,ARM] Select one of KVM/arm64's modes of operation. 2362 2363 none: Forcefully disable KVM. 2364 2365 nvhe: Standard nVHE-based mode, without support for 2366 protected guests. 2367 2368 protected: nVHE-based mode with support for guests whose 2369 state is kept private from the host. 2370 Not valid if the kernel is running in EL2. 2371 2372 Defaults to VHE/nVHE based on hardware support and 2373 the value of CONFIG_ARM64_VHE. 2374 2375 kvm-arm.vgic_v3_group0_trap= 2376 [KVM,ARM] Trap guest accesses to GICv3 group-0 2377 system registers 2378 2379 kvm-arm.vgic_v3_group1_trap= 2380 [KVM,ARM] Trap guest accesses to GICv3 group-1 2381 system registers 2382 2383 kvm-arm.vgic_v3_common_trap= 2384 [KVM,ARM] Trap guest accesses to GICv3 common 2385 system registers 2386 2387 kvm-arm.vgic_v4_enable= 2388 [KVM,ARM] Allow use of GICv4 for direct injection of 2389 LPIs. 2390 2391 kvm_cma_resv_ratio=n [PPC] 2392 Reserves given percentage from system memory area for 2393 contiguous memory allocation for KVM hash pagetable 2394 allocation. 2395 By default it reserves 5% of total system memory. 2396 Format: <integer> 2397 Default: 5 2398 2399 kvm-intel.ept= [KVM,Intel] Disable extended page tables 2400 (virtualized MMU) support on capable Intel chips. 2401 Default is 1 (enabled) 2402 2403 kvm-intel.emulate_invalid_guest_state= 2404 [KVM,Intel] Disable emulation of invalid guest state. 2405 Ignored if kvm-intel.enable_unrestricted_guest=1, as 2406 guest state is never invalid for unrestricted guests. 2407 This param doesn't apply to nested guests (L2), as KVM 2408 never emulates invalid L2 guest state. 2409 Default is 1 (enabled) 2410 2411 kvm-intel.flexpriority= 2412 [KVM,Intel] Disable FlexPriority feature (TPR shadow). 2413 Default is 1 (enabled) 2414 2415 kvm-intel.nested= 2416 [KVM,Intel] Enable VMX nesting (nVMX). 2417 Default is 0 (disabled) 2418 2419 kvm-intel.unrestricted_guest= 2420 [KVM,Intel] Disable unrestricted guest feature 2421 (virtualized real and unpaged mode) on capable 2422 Intel chips. Default is 1 (enabled) 2423 2424 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault 2425 CVE-2018-3620. 2426 2427 Valid arguments: never, cond, always 2428 2429 always: L1D cache flush on every VMENTER. 2430 cond: Flush L1D on VMENTER only when the code between 2431 VMEXIT and VMENTER can leak host memory. 2432 never: Disables the mitigation 2433 2434 Default is cond (do L1 cache flush in specific instances) 2435 2436 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification 2437 feature (tagged TLBs) on capable Intel chips. 2438 Default is 1 (enabled) 2439 2440 l1tf= [X86] Control mitigation of the L1TF vulnerability on 2441 affected CPUs 2442 2443 The kernel PTE inversion protection is unconditionally 2444 enabled and cannot be disabled. 2445 2446 full 2447 Provides all available mitigations for the 2448 L1TF vulnerability. Disables SMT and 2449 enables all mitigations in the 2450 hypervisors, i.e. unconditional L1D flush. 2451 2452 SMT control and L1D flush control via the 2453 sysfs interface is still possible after 2454 boot. Hypervisors will issue a warning 2455 when the first VM is started in a 2456 potentially insecure configuration, 2457 i.e. SMT enabled or L1D flush disabled. 2458 2459 full,force 2460 Same as 'full', but disables SMT and L1D 2461 flush runtime control. Implies the 2462 'nosmt=force' command line option. 2463 (i.e. sysfs control of SMT is disabled.) 2464 2465 flush 2466 Leaves SMT enabled and enables the default 2467 hypervisor mitigation, i.e. conditional 2468 L1D flush. 2469 2470 SMT control and L1D flush control via the 2471 sysfs interface is still possible after 2472 boot. Hypervisors will issue a warning 2473 when the first VM is started in a 2474 potentially insecure configuration, 2475 i.e. SMT enabled or L1D flush disabled. 2476 2477 flush,nosmt 2478 2479 Disables SMT and enables the default 2480 hypervisor mitigation. 2481 2482 SMT control and L1D flush control via the 2483 sysfs interface is still possible after 2484 boot. Hypervisors will issue a warning 2485 when the first VM is started in a 2486 potentially insecure configuration, 2487 i.e. SMT enabled or L1D flush disabled. 2488 2489 flush,nowarn 2490 Same as 'flush', but hypervisors will not 2491 warn when a VM is started in a potentially 2492 insecure configuration. 2493 2494 off 2495 Disables hypervisor mitigations and doesn't 2496 emit any warnings. 2497 It also drops the swap size and available 2498 RAM limit restriction on both hypervisor and 2499 bare metal. 2500 2501 Default is 'flush'. 2502 2503 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst 2504 2505 l2cr= [PPC] 2506 2507 l3cr= [PPC] 2508 2509 lapic [X86-32,APIC] Enable the local APIC even if BIOS 2510 disabled it. 2511 2512 lapic= [X86,APIC] Do not use TSC deadline 2513 value for LAPIC timer one-shot implementation. Default 2514 back to the programmable timer unit in the LAPIC. 2515 Format: notscdeadline 2516 2517 lapic_timer_c2_ok [X86,APIC] trust the local apic timer 2518 in C2 power state. 2519 2520 libata.dma= [LIBATA] DMA control 2521 libata.dma=0 Disable all PATA and SATA DMA 2522 libata.dma=1 PATA and SATA Disk DMA only 2523 libata.dma=2 ATAPI (CDROM) DMA only 2524 libata.dma=4 Compact Flash DMA only 2525 Combinations also work, so libata.dma=3 enables DMA 2526 for disks and CDROMs, but not CFs. 2527 2528 libata.ignore_hpa= [LIBATA] Ignore HPA limit 2529 libata.ignore_hpa=0 keep BIOS limits (default) 2530 libata.ignore_hpa=1 ignore limits, using full disk 2531 2532 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume 2533 when set. 2534 Format: <int> 2535 2536 libata.force= [LIBATA] Force configurations. The format is comma 2537 separated list of "[ID:]VAL" where ID is 2538 PORT[.DEVICE]. PORT and DEVICE are decimal numbers 2539 matching port, link or device. Basically, it matches 2540 the ATA ID string printed on console by libata. If 2541 the whole ID part is omitted, the last PORT and DEVICE 2542 values are used. If ID hasn't been specified yet, the 2543 configuration applies to all ports, links and devices. 2544 2545 If only DEVICE is omitted, the parameter applies to 2546 the port and all links and devices behind it. DEVICE 2547 number of 0 either selects the first device or the 2548 first fan-out link behind PMP device. It does not 2549 select the host link. DEVICE number of 15 selects the 2550 host link and device attached to it. 2551 2552 The VAL specifies the configuration to force. As long 2553 as there's no ambiguity shortcut notation is allowed. 2554 For example, both 1.5 and 1.5G would work for 1.5Gbps. 2555 The following configurations can be forced. 2556 2557 * Cable type: 40c, 80c, short40c, unk, ign or sata. 2558 Any ID with matching PORT is used. 2559 2560 * SATA link speed limit: 1.5Gbps or 3.0Gbps. 2561 2562 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7]. 2563 udma[/][16,25,33,44,66,100,133] notation is also 2564 allowed. 2565 2566 * [no]ncq: Turn on or off NCQ. 2567 2568 * [no]ncqtrim: Turn off queued DSM TRIM. 2569 2570 * nohrst, nosrst, norst: suppress hard, soft 2571 and both resets. 2572 2573 * rstonce: only attempt one reset during 2574 hot-unplug link recovery 2575 2576 * dump_id: dump IDENTIFY data. 2577 2578 * atapi_dmadir: Enable ATAPI DMADIR bridge support 2579 2580 * disable: Disable this device. 2581 2582 If there are multiple matching configurations changing 2583 the same attribute, the last one is used. 2584 2585 memblock=debug [KNL] Enable memblock debug messages. 2586 2587 android12_only.will_be_removed_soon.memblock_nomap_remove= [KNL] 2588 Setting this to true through kernel command line will 2589 call memblock_remove on the regions marked with no-map 2590 property thereby saving memory by removing page structs 2591 for those regions. By default this is set to false. 2592 2593 load_ramdisk= [RAM] [Deprecated] 2594 2595 lockd.nlm_grace_period=P [NFS] Assign grace period. 2596 Format: <integer> 2597 2598 lockd.nlm_tcpport=N [NFS] Assign TCP port. 2599 Format: <integer> 2600 2601 lockd.nlm_timeout=T [NFS] Assign timeout value. 2602 Format: <integer> 2603 2604 lockd.nlm_udpport=M [NFS] Assign UDP port. 2605 Format: <integer> 2606 2607 lockdown= [SECURITY] 2608 { integrity | confidentiality } 2609 Enable the kernel lockdown feature. If set to 2610 integrity, kernel features that allow userland to 2611 modify the running kernel are disabled. If set to 2612 confidentiality, kernel features that allow userland 2613 to extract confidential information from the kernel 2614 are also disabled. 2615 2616 locktorture.nreaders_stress= [KNL] 2617 Set the number of locking read-acquisition kthreads. 2618 Defaults to being automatically set based on the 2619 number of online CPUs. 2620 2621 locktorture.nwriters_stress= [KNL] 2622 Set the number of locking write-acquisition kthreads. 2623 2624 locktorture.onoff_holdoff= [KNL] 2625 Set time (s) after boot for CPU-hotplug testing. 2626 2627 locktorture.onoff_interval= [KNL] 2628 Set time (s) between CPU-hotplug operations, or 2629 zero to disable CPU-hotplug testing. 2630 2631 locktorture.shuffle_interval= [KNL] 2632 Set task-shuffle interval (jiffies). Shuffling 2633 tasks allows some CPUs to go into dyntick-idle 2634 mode during the locktorture test. 2635 2636 locktorture.shutdown_secs= [KNL] 2637 Set time (s) after boot system shutdown. This 2638 is useful for hands-off automated testing. 2639 2640 locktorture.stat_interval= [KNL] 2641 Time (s) between statistics printk()s. 2642 2643 locktorture.stutter= [KNL] 2644 Time (s) to stutter testing, for example, 2645 specifying five seconds causes the test to run for 2646 five seconds, wait for five seconds, and so on. 2647 This tests the locking primitive's ability to 2648 transition abruptly to and from idle. 2649 2650 locktorture.torture_type= [KNL] 2651 Specify the locking implementation to test. 2652 2653 locktorture.verbose= [KNL] 2654 Enable additional printk() statements. 2655 2656 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver 2657 Format: <irq> 2658 2659 loglevel= All Kernel Messages with a loglevel smaller than the 2660 console loglevel will be printed to the console. It can 2661 also be changed with klogd or other programs. The 2662 loglevels are defined as follows: 2663 2664 0 (KERN_EMERG) system is unusable 2665 1 (KERN_ALERT) action must be taken immediately 2666 2 (KERN_CRIT) critical conditions 2667 3 (KERN_ERR) error conditions 2668 4 (KERN_WARNING) warning conditions 2669 5 (KERN_NOTICE) normal but significant condition 2670 6 (KERN_INFO) informational 2671 7 (KERN_DEBUG) debug-level messages 2672 2673 log_buf_len=n[KMG] Sets the size of the printk ring buffer, 2674 in bytes. n must be a power of two and greater 2675 than the minimal size. The minimal size is defined 2676 by LOG_BUF_SHIFT kernel config parameter. There is 2677 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter 2678 that allows to increase the default size depending on 2679 the number of CPUs. See init/Kconfig for more details. 2680 2681 logo.nologo [FB] Disables display of the built-in Linux logo. 2682 This may be used to provide more screen space for 2683 kernel log messages and is useful when debugging 2684 kernel boot problems. 2685 2686 lp=0 [LP] Specify parallel ports to use, e.g, 2687 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses 2688 lp=reset first parallel port). 'lp=0' disables the 2689 lp=auto printer driver. 'lp=reset' (which can be 2690 specified in addition to the ports) causes 2691 attached printers to be reset. Using 2692 lp=port1,port2,... specifies the parallel ports 2693 to associate lp devices with, starting with 2694 lp0. A port specification may be 'none' to skip 2695 that lp device, or a parport name such as 2696 'parport0'. Specifying 'lp=auto' instead of a 2697 port specification list means that device IDs 2698 from each port should be examined, to see if 2699 an IEEE 1284-compliant printer is attached; if 2700 so, the driver will manage that printer. 2701 See also header of drivers/char/lp.c. 2702 2703 lpj=n [KNL] 2704 Sets loops_per_jiffy to given constant, thus avoiding 2705 time-consuming boot-time autodetection (up to 250 ms per 2706 CPU). 0 enables autodetection (default). To determine 2707 the correct value for your kernel, boot with normal 2708 autodetection and see what value is printed. Note that 2709 on SMP systems the preset will be applied to all CPUs, 2710 which is likely to cause problems if your CPUs need 2711 significantly divergent settings. An incorrect value 2712 will cause delays in the kernel to be wrong, leading to 2713 unpredictable I/O errors and other breakage. Although 2714 unlikely, in the extreme case this might damage your 2715 hardware. 2716 2717 ltpc= [NET] 2718 Format: <io>,<irq>,<dma> 2719 2720 lsm.debug [SECURITY] Enable LSM initialization debugging output. 2721 2722 lsm=lsm1,...,lsmN 2723 [SECURITY] Choose order of LSM initialization. This 2724 overrides CONFIG_LSM, and the "security=" parameter. 2725 2726 machvec= [IA-64] Force the use of a particular machine-vector 2727 (machvec) in a generic kernel. 2728 Example: machvec=hpzx1 2729 2730 machtype= [Loongson] Share the same kernel image file between 2731 different yeeloong laptops. 2732 Example: machtype=lemote-yeeloong-2f-7inch 2733 2734 max_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory greater 2735 than or equal to this physical address is ignored. 2736 2737 maxcpus= [SMP] Maximum number of processors that an SMP kernel 2738 will bring up during bootup. maxcpus=n : n >= 0 limits 2739 the kernel to bring up 'n' processors. Surely after 2740 bootup you can bring up the other plugged cpu by executing 2741 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus 2742 only takes effect during system bootup. 2743 While n=0 is a special case, it is equivalent to "nosmp", 2744 which also disables the IO APIC. 2745 2746 max_loop= [LOOP] The number of loop block devices that get 2747 (loop.max_loop) unconditionally pre-created at init time. The default 2748 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead 2749 of statically allocating a predefined number, loop 2750 devices can be requested on-demand with the 2751 /dev/loop-control interface. 2752 2753 mce [X86-32] Machine Check Exception 2754 2755 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.rst 2756 2757 md= [HW] RAID subsystems devices and level 2758 See Documentation/admin-guide/md.rst. 2759 2760 mdacon= [MDA] 2761 Format: <first>,<last> 2762 Specifies range of consoles to be captured by the MDA. 2763 2764 mds= [X86,INTEL] 2765 Control mitigation for the Micro-architectural Data 2766 Sampling (MDS) vulnerability. 2767 2768 Certain CPUs are vulnerable to an exploit against CPU 2769 internal buffers which can forward information to a 2770 disclosure gadget under certain conditions. 2771 2772 In vulnerable processors, the speculatively 2773 forwarded data can be used in a cache side channel 2774 attack, to access data to which the attacker does 2775 not have direct access. 2776 2777 This parameter controls the MDS mitigation. The 2778 options are: 2779 2780 full - Enable MDS mitigation on vulnerable CPUs 2781 full,nosmt - Enable MDS mitigation and disable 2782 SMT on vulnerable CPUs 2783 off - Unconditionally disable MDS mitigation 2784 2785 On TAA-affected machines, mds=off can be prevented by 2786 an active TAA mitigation as both vulnerabilities are 2787 mitigated with the same mechanism so in order to disable 2788 this mitigation, you need to specify tsx_async_abort=off 2789 too. 2790 2791 Not specifying this option is equivalent to 2792 mds=full. 2793 2794 For details see: Documentation/admin-guide/hw-vuln/mds.rst 2795 2796 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory 2797 Amount of memory to be used in cases as follows: 2798 2799 1 for test; 2800 2 when the kernel is not able to see the whole system memory; 2801 3 memory that lies after 'mem=' boundary is excluded from 2802 the hypervisor, then assigned to KVM guests. 2803 2804 [X86] Work as limiting max address. Use together 2805 with memmap= to avoid physical address space collisions. 2806 Without memmap= PCI devices could be placed at addresses 2807 belonging to unused RAM. 2808 2809 Note that this only takes effects during boot time since 2810 in above case 3, memory may need be hot added after boot 2811 if system memory of hypervisor is not sufficient. 2812 2813 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel 2814 memory. 2815 2816 memchunk=nn[KMG] 2817 [KNL,SH] Allow user to override the default size for 2818 per-device physically contiguous DMA buffers. 2819 2820 memhp_default_state=online/offline 2821 [KNL] Set the initial state for the memory hotplug 2822 onlining policy. If not specified, the default value is 2823 set according to the 2824 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config 2825 option. 2826 See Documentation/admin-guide/mm/memory-hotplug.rst. 2827 2828 memmap=exactmap [KNL,X86] Enable setting of an exact 2829 E820 memory map, as specified by the user. 2830 Such memmap=exactmap lines can be constructed based on 2831 BIOS output or other requirements. See the memmap=nn@ss 2832 option description. 2833 2834 memmap=nn[KMG]@ss[KMG] 2835 [KNL] Force usage of a specific region of memory. 2836 Region of memory to be used is from ss to ss+nn. 2837 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG], 2838 which limits max address to nn[KMG]. 2839 Multiple different regions can be specified, 2840 comma delimited. 2841 Example: 2842 memmap=100M@2G,100M#3G,1G!1024G 2843 2844 memmap=nn[KMG]#ss[KMG] 2845 [KNL,ACPI] Mark specific memory as ACPI data. 2846 Region of memory to be marked is from ss to ss+nn. 2847 2848 memmap=nn[KMG]$ss[KMG] 2849 [KNL,ACPI] Mark specific memory as reserved. 2850 Region of memory to be reserved is from ss to ss+nn. 2851 Example: Exclude memory from 0x18690000-0x1869ffff 2852 memmap=64K$0x18690000 2853 or 2854 memmap=0x10000$0x18690000 2855 Some bootloaders may need an escape character before '$', 2856 like Grub2, otherwise '$' and the following number 2857 will be eaten. 2858 2859 memmap=nn[KMG]!ss[KMG] 2860 [KNL,X86] Mark specific memory as protected. 2861 Region of memory to be used, from ss to ss+nn. 2862 The memory region may be marked as e820 type 12 (0xc) 2863 and is NVDIMM or ADR memory. 2864 2865 memmap=<size>%<offset>-<oldtype>+<newtype> 2866 [KNL,ACPI] Convert memory within the specified region 2867 from <oldtype> to <newtype>. If "-<oldtype>" is left 2868 out, the whole region will be marked as <newtype>, 2869 even if previously unavailable. If "+<newtype>" is left 2870 out, matching memory will be removed. Types are 2871 specified as e820 types, e.g., 1 = RAM, 2 = reserved, 2872 3 = ACPI, 12 = PRAM. 2873 2874 memory_corruption_check=0/1 [X86] 2875 Some BIOSes seem to corrupt the first 64k of 2876 memory when doing things like suspend/resume. 2877 Setting this option will scan the memory 2878 looking for corruption. Enabling this will 2879 both detect corruption and prevent the kernel 2880 from using the memory being corrupted. 2881 However, its intended as a diagnostic tool; if 2882 repeatable BIOS-originated corruption always 2883 affects the same memory, you can use memmap= 2884 to prevent the kernel from using that memory. 2885 2886 memory_corruption_check_size=size [X86] 2887 By default it checks for corruption in the low 2888 64k, making this memory unavailable for normal 2889 use. Use this parameter to scan for 2890 corruption in more or less memory. 2891 2892 memory_corruption_check_period=seconds [X86] 2893 By default it checks for corruption every 60 2894 seconds. Use this parameter to check at some 2895 other rate. 0 disables periodic checking. 2896 2897 memtest= [KNL,X86,ARM,PPC] Enable memtest 2898 Format: <integer> 2899 default : 0 <disable> 2900 Specifies the number of memtest passes to be 2901 performed. Each pass selects another test 2902 pattern from a given set of patterns. Memtest 2903 fills the memory with this pattern, validates 2904 memory contents and reserves bad memory 2905 regions that are detected. 2906 2907 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control 2908 Valid arguments: on, off 2909 Default (depends on kernel configuration option): 2910 on (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=y) 2911 off (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=n) 2912 mem_encrypt=on: Activate SME 2913 mem_encrypt=off: Do not activate SME 2914 2915 Refer to Documentation/virt/kvm/amd-memory-encryption.rst 2916 for details on when memory encryption can be activated. 2917 2918 mem_sleep_default= [SUSPEND] Default system suspend mode: 2919 s2idle - Suspend-To-Idle 2920 shallow - Power-On Suspend or equivalent (if supported) 2921 deep - Suspend-To-RAM or equivalent (if supported) 2922 See Documentation/admin-guide/pm/sleep-states.rst. 2923 2924 meye.*= [HW] Set MotionEye Camera parameters 2925 See Documentation/admin-guide/media/meye.rst. 2926 2927 mfgpt_irq= [IA-32] Specify the IRQ to use for the 2928 Multi-Function General Purpose Timers on AMD Geode 2929 platforms. 2930 2931 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when 2932 the BIOS has incorrectly applied a workaround. TinyBIOS 2933 version 0.98 is known to be affected, 0.99 fixes the 2934 problem by letting the user disable the workaround. 2935 2936 mga= [HW,DRM] 2937 2938 min_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory below this 2939 physical address is ignored. 2940 2941 mini2440= [ARM,HW,KNL] 2942 Format:[0..2][b][c][t] 2943 Default: "0tb" 2944 MINI2440 configuration specification: 2945 0 - The attached screen is the 3.5" TFT 2946 1 - The attached screen is the 7" TFT 2947 2 - The VGA Shield is attached (1024x768) 2948 Leaving out the screen size parameter will not load 2949 the TFT driver, and the framebuffer will be left 2950 unconfigured. 2951 b - Enable backlight. The TFT backlight pin will be 2952 linked to the kernel VESA blanking code and a GPIO 2953 LED. This parameter is not necessary when using the 2954 VGA shield. 2955 c - Enable the s3c camera interface. 2956 t - Reserved for enabling touchscreen support. The 2957 touchscreen support is not enabled in the mainstream 2958 kernel as of 2.6.30, a preliminary port can be found 2959 in the "bleeding edge" mini2440 support kernel at 2960 https://repo.or.cz/w/linux-2.6/mini2440.git 2961 2962 mitigations= 2963 [X86,PPC,S390,ARM64] Control optional mitigations for 2964 CPU vulnerabilities. This is a set of curated, 2965 arch-independent options, each of which is an 2966 aggregation of existing arch-specific options. 2967 2968 off 2969 Disable all optional CPU mitigations. This 2970 improves system performance, but it may also 2971 expose users to several CPU vulnerabilities. 2972 Equivalent to: gather_data_sampling=off [X86] 2973 kpti=0 [ARM64] 2974 kvm.nx_huge_pages=off [X86] 2975 l1tf=off [X86] 2976 mds=off [X86] 2977 mmio_stale_data=off [X86] 2978 no_entry_flush [PPC] 2979 no_uaccess_flush [PPC] 2980 nobp=0 [S390] 2981 nopti [X86,PPC] 2982 nospectre_v1 [X86,PPC] 2983 nospectre_v2 [X86,PPC,S390,ARM64] 2984 retbleed=off [X86] 2985 spec_store_bypass_disable=off [X86,PPC] 2986 spectre_v2_user=off [X86] 2987 ssbd=force-off [ARM64] 2988 tsx_async_abort=off [X86] 2989 2990 Exceptions: 2991 This does not have any effect on 2992 kvm.nx_huge_pages when 2993 kvm.nx_huge_pages=force. 2994 2995 auto (default) 2996 Mitigate all CPU vulnerabilities, but leave SMT 2997 enabled, even if it's vulnerable. This is for 2998 users who don't want to be surprised by SMT 2999 getting disabled across kernel upgrades, or who 3000 have other ways of avoiding SMT-based attacks. 3001 Equivalent to: (default behavior) 3002 3003 auto,nosmt 3004 Mitigate all CPU vulnerabilities, disabling SMT 3005 if needed. This is for users who always want to 3006 be fully mitigated, even if it means losing SMT. 3007 Equivalent to: l1tf=flush,nosmt [X86] 3008 mds=full,nosmt [X86] 3009 tsx_async_abort=full,nosmt [X86] 3010 mmio_stale_data=full,nosmt [X86] 3011 retbleed=auto,nosmt [X86] 3012 3013 mminit_loglevel= 3014 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this 3015 parameter allows control of the logging verbosity for 3016 the additional memory initialisation checks. A value 3017 of 0 disables mminit logging and a level of 4 will 3018 log everything. Information is printed at KERN_DEBUG 3019 so loglevel=8 may also need to be specified. 3020 3021 mmio_stale_data= 3022 [X86,INTEL] Control mitigation for the Processor 3023 MMIO Stale Data vulnerabilities. 3024 3025 Processor MMIO Stale Data is a class of 3026 vulnerabilities that may expose data after an MMIO 3027 operation. Exposed data could originate or end in 3028 the same CPU buffers as affected by MDS and TAA. 3029 Therefore, similar to MDS and TAA, the mitigation 3030 is to clear the affected CPU buffers. 3031 3032 This parameter controls the mitigation. The 3033 options are: 3034 3035 full - Enable mitigation on vulnerable CPUs 3036 3037 full,nosmt - Enable mitigation and disable SMT on 3038 vulnerable CPUs. 3039 3040 off - Unconditionally disable mitigation 3041 3042 On MDS or TAA affected machines, 3043 mmio_stale_data=off can be prevented by an active 3044 MDS or TAA mitigation as these vulnerabilities are 3045 mitigated with the same mechanism so in order to 3046 disable this mitigation, you need to specify 3047 mds=off and tsx_async_abort=off too. 3048 3049 Not specifying this option is equivalent to 3050 mmio_stale_data=full. 3051 3052 For details see: 3053 Documentation/admin-guide/hw-vuln/processor_mmio_stale_data.rst 3054 3055 module.sig_enforce 3056 [KNL] When CONFIG_MODULE_SIG is set, this means that 3057 modules without (valid) signatures will fail to load. 3058 Note that if CONFIG_MODULE_SIG_FORCE is set, that 3059 is always true, so this option does nothing. 3060 3061 module_blacklist= [KNL] Do not load a comma-separated list of 3062 modules. Useful for debugging problem modules. 3063 3064 mousedev.tap_time= 3065 [MOUSE] Maximum time between finger touching and 3066 leaving touchpad surface for touch to be considered 3067 a tap and be reported as a left button click (for 3068 touchpads working in absolute mode only). 3069 Format: <msecs> 3070 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices 3071 reporting absolute coordinates, such as tablets 3072 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices 3073 reporting absolute coordinates, such as tablets 3074 3075 movablecore= [KNL,X86,IA-64,PPC] 3076 Format: nn[KMGTPE] | nn% 3077 This parameter is the complement to kernelcore=, it 3078 specifies the amount of memory used for migratable 3079 allocations. If both kernelcore and movablecore is 3080 specified, then kernelcore will be at *least* the 3081 specified value but may be more. If movablecore on its 3082 own is specified, the administrator must be careful 3083 that the amount of memory usable for all allocations 3084 is not too small. 3085 3086 movable_node [KNL] Boot-time switch to make hotplugable memory 3087 NUMA nodes to be movable. This means that the memory 3088 of such nodes will be usable only for movable 3089 allocations which rules out almost all kernel 3090 allocations. Use with caution! 3091 3092 MTD_Partition= [MTD] 3093 Format: <name>,<region-number>,<size>,<offset> 3094 3095 MTD_Region= [MTD] Format: 3096 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>] 3097 3098 mtdparts= [MTD] 3099 See drivers/mtd/parsers/cmdlinepart.c 3100 3101 multitce=off [PPC] This parameter disables the use of the pSeries 3102 firmware feature for updating multiple TCE entries 3103 at a time. 3104 3105 kswapd_per_node= 3106 kswapd_per_node allows you to control the number of kswapd threads 3107 running on the system. This provides the ability to devote additional 3108 CPU resources toward proactive page replacement with the goal of 3109 reducing direct reclaims. When direct reclaims are prevented, the CPU 3110 consumed by them is prevented as well. Depending on the workload, the 3111 result can cause aggregate CPU usage on the system to go up, down or 3112 stay the same. 3113 3114 More aggressive page replacement can reduce direct reclaims which 3115 cause latency for tasks and decrease throughput when doing filesystem 3116 IO through the pagecache. Direct reclaims are recorded using the 3117 allocstall counter in /proc/vmstat. 3118 3119 The range of acceptible values are 1-16. Always start with lower 3120 values in the 2-6 range. Higher values should be justified with 3121 testing. If direct reclaims occur in spite of high values, the cost 3122 of direct reclaims (in latency) that occur can be higher due to 3123 increased lock contention. 3124 3125 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration 3126 3127 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock] 3128 3129 boundary - index of last SLC block on Flex-OneNAND. 3130 The remaining blocks are configured as MLC blocks. 3131 lock - Configure if Flex-OneNAND boundary should be locked. 3132 Once locked, the boundary cannot be changed. 3133 1 indicates lock status, 0 indicates unlock status. 3134 3135 mtdset= [ARM] 3136 ARM/S3C2412 JIVE boot control 3137 3138 See arch/arm/mach-s3c2412/mach-jive.c 3139 3140 mtouchusb.raw_coordinates= 3141 [HW] Make the MicroTouch USB driver use raw coordinates 3142 ('y', default) or cooked coordinates ('n') 3143 3144 mtrr_chunk_size=nn[KMG] [X86] 3145 used for mtrr cleanup. It is largest continuous chunk 3146 that could hold holes aka. UC entries. 3147 3148 mtrr_gran_size=nn[KMG] [X86] 3149 Used for mtrr cleanup. It is granularity of mtrr block. 3150 Default is 1. 3151 Large value could prevent small alignment from 3152 using up MTRRs. 3153 3154 mtrr_spare_reg_nr=n [X86] 3155 Format: <integer> 3156 Range: 0,7 : spare reg number 3157 Default : 1 3158 Used for mtrr cleanup. It is spare mtrr entries number. 3159 Set to 2 or more if your graphical card needs more. 3160 3161 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card 3162 3163 netdev= [NET] Network devices parameters 3164 Format: <irq>,<io>,<mem_start>,<mem_end>,<name> 3165 Note that mem_start is often overloaded to mean 3166 something different and driver-specific. 3167 This usage is only documented in each driver source 3168 file if at all. 3169 3170 nf_conntrack.acct= 3171 [NETFILTER] Enable connection tracking flow accounting 3172 0 to disable accounting 3173 1 to enable accounting 3174 Default value is 0. 3175 3176 nfsaddrs= [NFS] Deprecated. Use ip= instead. 3177 See Documentation/admin-guide/nfs/nfsroot.rst. 3178 3179 nfsroot= [NFS] nfs root filesystem for disk-less boxes. 3180 See Documentation/admin-guide/nfs/nfsroot.rst. 3181 3182 nfsrootdebug [NFS] enable nfsroot debugging messages. 3183 See Documentation/admin-guide/nfs/nfsroot.rst. 3184 3185 nfs.callback_nr_threads= 3186 [NFSv4] set the total number of threads that the 3187 NFS client will assign to service NFSv4 callback 3188 requests. 3189 3190 nfs.callback_tcpport= 3191 [NFS] set the TCP port on which the NFSv4 callback 3192 channel should listen. 3193 3194 nfs.cache_getent= 3195 [NFS] sets the pathname to the program which is used 3196 to update the NFS client cache entries. 3197 3198 nfs.cache_getent_timeout= 3199 [NFS] sets the timeout after which an attempt to 3200 update a cache entry is deemed to have failed. 3201 3202 nfs.idmap_cache_timeout= 3203 [NFS] set the maximum lifetime for idmapper cache 3204 entries. 3205 3206 nfs.enable_ino64= 3207 [NFS] enable 64-bit inode numbers. 3208 If zero, the NFS client will fake up a 32-bit inode 3209 number for the readdir() and stat() syscalls instead 3210 of returning the full 64-bit number. 3211 The default is to return 64-bit inode numbers. 3212 3213 nfs.max_session_cb_slots= 3214 [NFSv4.1] Sets the maximum number of session 3215 slots the client will assign to the callback 3216 channel. This determines the maximum number of 3217 callbacks the client will process in parallel for 3218 a particular server. 3219 3220 nfs.max_session_slots= 3221 [NFSv4.1] Sets the maximum number of session slots 3222 the client will attempt to negotiate with the server. 3223 This limits the number of simultaneous RPC requests 3224 that the client can send to the NFSv4.1 server. 3225 Note that there is little point in setting this 3226 value higher than the max_tcp_slot_table_limit. 3227 3228 nfs.nfs4_disable_idmapping= 3229 [NFSv4] When set to the default of '1', this option 3230 ensures that both the RPC level authentication 3231 scheme and the NFS level operations agree to use 3232 numeric uids/gids if the mount is using the 3233 'sec=sys' security flavour. In effect it is 3234 disabling idmapping, which can make migration from 3235 legacy NFSv2/v3 systems to NFSv4 easier. 3236 Servers that do not support this mode of operation 3237 will be autodetected by the client, and it will fall 3238 back to using the idmapper. 3239 To turn off this behaviour, set the value to '0'. 3240 nfs.nfs4_unique_id= 3241 [NFS4] Specify an additional fixed unique ident- 3242 ification string that NFSv4 clients can insert into 3243 their nfs_client_id4 string. This is typically a 3244 UUID that is generated at system install time. 3245 3246 nfs.send_implementation_id = 3247 [NFSv4.1] Send client implementation identification 3248 information in exchange_id requests. 3249 If zero, no implementation identification information 3250 will be sent. 3251 The default is to send the implementation identification 3252 information. 3253 3254 nfs.recover_lost_locks = 3255 [NFSv4] Attempt to recover locks that were lost due 3256 to a lease timeout on the server. Please note that 3257 doing this risks data corruption, since there are 3258 no guarantees that the file will remain unchanged 3259 after the locks are lost. 3260 If you want to enable the kernel legacy behaviour of 3261 attempting to recover these locks, then set this 3262 parameter to '1'. 3263 The default parameter value of '0' causes the kernel 3264 not to attempt recovery of lost locks. 3265 3266 nfs4.layoutstats_timer = 3267 [NFSv4.2] Change the rate at which the kernel sends 3268 layoutstats to the pNFS metadata server. 3269 3270 Setting this to value to 0 causes the kernel to use 3271 whatever value is the default set by the layout 3272 driver. A non-zero value sets the minimum interval 3273 in seconds between layoutstats transmissions. 3274 3275 nfsd.nfs4_disable_idmapping= 3276 [NFSv4] When set to the default of '1', the NFSv4 3277 server will return only numeric uids and gids to 3278 clients using auth_sys, and will accept numeric uids 3279 and gids from such clients. This is intended to ease 3280 migration from NFSv2/v3. 3281 3282 nmi_backtrace.backtrace_idle [KNL] 3283 Dump stacks even of idle CPUs in response to an 3284 NMI stack-backtrace request. 3285 3286 nmi_debug= [KNL,SH] Specify one or more actions to take 3287 when a NMI is triggered. 3288 Format: [state][,regs][,debounce][,die] 3289 3290 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels 3291 Format: [panic,][nopanic,][num] 3292 Valid num: 0 or 1 3293 0 - turn hardlockup detector in nmi_watchdog off 3294 1 - turn hardlockup detector in nmi_watchdog on 3295 When panic is specified, panic when an NMI watchdog 3296 timeout occurs (or 'nopanic' to not panic on an NMI 3297 watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set) 3298 To disable both hard and soft lockup detectors, 3299 please see 'nowatchdog'. 3300 This is useful when you use a panic=... timeout and 3301 need the box quickly up again. 3302 3303 These settings can be accessed at runtime via 3304 the nmi_watchdog and hardlockup_panic sysctls. 3305 3306 netpoll.carrier_timeout= 3307 [NET] Specifies amount of time (in seconds) that 3308 netpoll should wait for a carrier. By default netpoll 3309 waits 4 seconds. 3310 3311 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths 3312 emulation library even if a 387 maths coprocessor 3313 is present. 3314 3315 no5lvl [X86-64] Disable 5-level paging mode. Forces 3316 kernel to use 4-level paging instead. 3317 3318 nofsgsbase [X86] Disables FSGSBASE instructions. 3319 3320 no_console_suspend 3321 [HW] Never suspend the console 3322 Disable suspending of consoles during suspend and 3323 hibernate operations. Once disabled, debugging 3324 messages can reach various consoles while the rest 3325 of the system is being put to sleep (ie, while 3326 debugging driver suspend/resume hooks). This may 3327 not work reliably with all consoles, but is known 3328 to work with serial and VGA consoles. 3329 To facilitate more flexible debugging, we also add 3330 console_suspend, a printk module parameter to control 3331 it. Users could use console_suspend (usually 3332 /sys/module/printk/parameters/console_suspend) to 3333 turn on/off it dynamically. 3334 3335 novmcoredd [KNL,KDUMP] 3336 Disable device dump. Device dump allows drivers to 3337 append dump data to vmcore so you can collect driver 3338 specified debug info. Drivers can append the data 3339 without any limit and this data is stored in memory, 3340 so this may cause significant memory stress. Disabling 3341 device dump can help save memory but the driver debug 3342 data will be no longer available. This parameter 3343 is only available when CONFIG_PROC_VMCORE_DEVICE_DUMP 3344 is set. 3345 3346 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien 3347 caches in the slab allocator. Saves per-node memory, 3348 but will impact performance. 3349 3350 noalign [KNL,ARM] 3351 3352 noaltinstr [S390] Disables alternative instructions patching 3353 (CPU alternatives feature). 3354 3355 noapic [SMP,APIC] Tells the kernel to not make use of any 3356 IOAPICs that may be present in the system. 3357 3358 noautogroup Disable scheduler automatic task group creation. 3359 3360 nobats [PPC] Do not use BATs for mapping kernel lowmem 3361 on "Classic" PPC cores. 3362 3363 nocache [ARM] 3364 3365 noclflush [BUGS=X86] Don't use the CLFLUSH instruction 3366 3367 nodelayacct [KNL] Disable per-task delay accounting 3368 3369 nodsp [SH] Disable hardware DSP at boot time. 3370 3371 noefi Disable EFI runtime services support. 3372 3373 no_entry_flush [PPC] Don't flush the L1-D cache when entering the kernel. 3374 3375 noexec [IA-64] 3376 3377 noexec [X86] 3378 On X86-32 available only on PAE configured kernels. 3379 noexec=on: enable non-executable mappings (default) 3380 noexec=off: disable non-executable mappings 3381 3382 nosmap [X86,PPC] 3383 Disable SMAP (Supervisor Mode Access Prevention) 3384 even if it is supported by processor. 3385 3386 nosmep [X86,PPC] 3387 Disable SMEP (Supervisor Mode Execution Prevention) 3388 even if it is supported by processor. 3389 3390 noexec32 [X86-64] 3391 This affects only 32-bit executables. 3392 noexec32=on: enable non-executable mappings (default) 3393 read doesn't imply executable mappings 3394 noexec32=off: disable non-executable mappings 3395 read implies executable mappings 3396 3397 nofpu [MIPS,SH] Disable hardware FPU at boot time. 3398 3399 nofxsr [BUGS=X86-32] Disables x86 floating point extended 3400 register save and restore. The kernel will only save 3401 legacy floating-point registers on task switch. 3402 3403 nohugeiomap [KNL,X86,PPC,ARM64] Disable kernel huge I/O mappings. 3404 3405 nosmt [KNL,S390] Disable symmetric multithreading (SMT). 3406 Equivalent to smt=1. 3407 3408 [KNL,X86] Disable symmetric multithreading (SMT). 3409 nosmt=force: Force disable SMT, cannot be undone 3410 via the sysfs control file. 3411 3412 nospectre_v1 [X86,PPC] Disable mitigations for Spectre Variant 1 3413 (bounds check bypass). With this option data leaks are 3414 possible in the system. 3415 3416 nospectre_v2 [X86,PPC_FSL_BOOK3E,ARM64] Disable all mitigations for 3417 the Spectre variant 2 (indirect branch prediction) 3418 vulnerability. System may allow data leaks with this 3419 option. 3420 3421 nospec_store_bypass_disable 3422 [HW] Disable all mitigations for the Speculative Store Bypass vulnerability 3423 3424 no_uaccess_flush 3425 [PPC] Don't flush the L1-D cache after accessing user data. 3426 3427 noxsave [BUGS=X86] Disables x86 extended register state save 3428 and restore using xsave. The kernel will fallback to 3429 enabling legacy floating-point and sse state. 3430 3431 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended 3432 register states. The kernel will fall back to use 3433 xsave to save the states. By using this parameter, 3434 performance of saving the states is degraded because 3435 xsave doesn't support modified optimization while 3436 xsaveopt supports it on xsaveopt enabled systems. 3437 3438 noxsaves [X86] Disables xsaves and xrstors used in saving and 3439 restoring x86 extended register state in compacted 3440 form of xsave area. The kernel will fall back to use 3441 xsaveopt and xrstor to save and restore the states 3442 in standard form of xsave area. By using this 3443 parameter, xsave area per process might occupy more 3444 memory on xsaves enabled systems. 3445 3446 nohlt [BUGS=ARM,SH] Tells the kernel that the sleep(SH) or 3447 wfi(ARM) instruction doesn't work correctly and not to 3448 use it. This is also useful when using JTAG debugger. 3449 3450 no_file_caps Tells the kernel not to honor file capabilities. The 3451 only way then for a file to be executed with privilege 3452 is to be setuid root or executed by root. 3453 3454 nohalt [IA-64] Tells the kernel not to use the power saving 3455 function PAL_HALT_LIGHT when idle. This increases 3456 power-consumption. On the positive side, it reduces 3457 interrupt wake-up latency, which may improve performance 3458 in certain environments such as networked servers or 3459 real-time systems. 3460 3461 no_hash_pointers 3462 Force pointers printed to the console or buffers to be 3463 unhashed. By default, when a pointer is printed via %p 3464 format string, that pointer is "hashed", i.e. obscured 3465 by hashing the pointer value. This is a security feature 3466 that hides actual kernel addresses from unprivileged 3467 users, but it also makes debugging the kernel more 3468 difficult since unequal pointers can no longer be 3469 compared. However, if this command-line option is 3470 specified, then all normal pointers will have their true 3471 value printed. This option should only be specified when 3472 debugging the kernel. Please do not use on production 3473 kernels. 3474 3475 nohibernate [HIBERNATION] Disable hibernation and resume. 3476 3477 nohz= [KNL] Boottime enable/disable dynamic ticks 3478 Valid arguments: on, off 3479 Default: on 3480 3481 nohz_full= [KNL,BOOT,SMP,ISOL] 3482 The argument is a cpu list, as described above. 3483 In kernels built with CONFIG_NO_HZ_FULL=y, set 3484 the specified list of CPUs whose tick will be stopped 3485 whenever possible. The boot CPU will be forced outside 3486 the range to maintain the timekeeping. Any CPUs 3487 in this list will have their RCU callbacks offloaded, 3488 just as if they had also been called out in the 3489 rcu_nocbs= boot parameter. 3490 3491 noiotrap [SH] Disables trapped I/O port accesses. 3492 3493 noirqdebug [X86-32] Disables the code which attempts to detect and 3494 disable unhandled interrupt sources. 3495 3496 no_timer_check [X86,APIC] Disables the code which tests for 3497 broken timer IRQ sources. 3498 3499 noisapnp [ISAPNP] Disables ISA PnP code. 3500 3501 noinitrd [RAM] Tells the kernel not to load any configured 3502 initial RAM disk. 3503 3504 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt 3505 remapping. 3506 [Deprecated - use intremap=off] 3507 3508 nointroute [IA-64] 3509 3510 noinvpcid [X86] Disable the INVPCID cpu feature. 3511 3512 nojitter [IA-64] Disables jitter checking for ITC timers. 3513 3514 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver 3515 3516 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page 3517 fault handling. 3518 3519 no-vmw-sched-clock 3520 [X86,PV_OPS] Disable paravirtualized VMware scheduler 3521 clock and use the default one. 3522 3523 no-steal-acc [X86,PV_OPS,ARM64] Disable paravirtualized steal time 3524 accounting. steal time is computed, but won't 3525 influence scheduler behaviour 3526 3527 nolapic [X86-32,APIC] Do not enable or use the local APIC. 3528 3529 nolapic_timer [X86-32,APIC] Do not use the local APIC timer. 3530 3531 noltlbs [PPC] Do not use large page/tlb entries for kernel 3532 lowmem mapping on PPC40x and PPC8xx 3533 3534 nomca [IA-64] Disable machine check abort handling 3535 3536 nomce [X86-32] Disable Machine Check Exception 3537 3538 nomfgpt [X86-32] Disable Multi-Function General Purpose 3539 Timer usage (for AMD Geode machines). 3540 3541 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to 3542 shutdown the other cpus. Instead use the REBOOT_VECTOR 3543 irq. 3544 3545 nomodule Disable module load 3546 3547 nopat [X86] Disable PAT (page attribute table extension of 3548 pagetables) support. 3549 3550 nopcid [X86-64] Disable the PCID cpu feature. 3551 3552 norandmaps Don't use address space randomization. Equivalent to 3553 echo 0 > /proc/sys/kernel/randomize_va_space 3554 3555 noreplace-smp [X86-32,SMP] Don't replace SMP instructions 3556 with UP alternatives 3557 3558 nordrand [X86] Disable kernel use of the RDRAND and 3559 RDSEED instructions even if they are supported 3560 by the processor. RDRAND and RDSEED are still 3561 available to user space applications. 3562 3563 noresume [SWSUSP] Disables resume and restores original swap 3564 space. 3565 3566 no-scroll [VGA] Disables scrollback. 3567 This is required for the Braillex ib80-piezo Braille 3568 reader made by F.H. Papenmeier (Germany). 3569 3570 nosbagart [IA-64] 3571 3572 nosep [BUGS=X86-32] Disables x86 SYSENTER/SYSEXIT support. 3573 3574 nosmp [SMP] Tells an SMP kernel to act as a UP kernel, 3575 and disable the IO APIC. legacy for "maxcpus=0". 3576 3577 nosoftlockup [KNL] Disable the soft-lockup detector. 3578 3579 nosync [HW,M68K] Disables sync negotiation for all devices. 3580 3581 nowatchdog [KNL] Disable both lockup detectors, i.e. 3582 soft-lockup and NMI watchdog (hard-lockup). 3583 3584 nowb [ARM] 3585 3586 nox2apic [X86-64,APIC] Do not enable x2APIC mode. 3587 3588 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when 3589 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off. 3590 Some features depend on CPU0. Known dependencies are: 3591 1. Resume from suspend/hibernate depends on CPU0. 3592 Suspend/hibernate will fail if CPU0 is offline and you 3593 need to online CPU0 before suspend/hibernate. 3594 2. PIC interrupts also depend on CPU0. CPU0 can't be 3595 removed if a PIC interrupt is detected. 3596 It's said poweroff/reboot may depend on CPU0 on some 3597 machines although I haven't seen such issues so far 3598 after CPU0 is offline on a few tested machines. 3599 If the dependencies are under your control, you can 3600 turn on cpu0_hotplug. 3601 3602 nps_mtm_hs_ctr= [KNL,ARC] 3603 This parameter sets the maximum duration, in 3604 cycles, each HW thread of the CTOP can run 3605 without interruptions, before HW switches it. 3606 The actual maximum duration is 16 times this 3607 parameter's value. 3608 Format: integer between 1 and 255 3609 Default: 255 3610 3611 nptcg= [IA-64] Override max number of concurrent global TLB 3612 purges which is reported from either PAL_VM_SUMMARY or 3613 SAL PALO. 3614 3615 nr_cpus= [SMP] Maximum number of processors that an SMP kernel 3616 could support. nr_cpus=n : n >= 1 limits the kernel to 3617 support 'n' processors. It could be larger than the 3618 number of already plugged CPU during bootup, later in 3619 runtime you can physically add extra cpu until it reaches 3620 n. So during boot up some boot time memory for per-cpu 3621 variables need be pre-allocated for later physical cpu 3622 hot plugging. 3623 3624 nr_uarts= [SERIAL] maximum number of UARTs to be registered. 3625 3626 numa_balancing= [KNL,X86] Enable or disable automatic NUMA balancing. 3627 Allowed values are enable and disable 3628 3629 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA. 3630 'node', 'default' can be specified 3631 This can be set from sysctl after boot. 3632 See Documentation/admin-guide/sysctl/vm.rst for details. 3633 3634 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver. 3635 See Documentation/core-api/debugging-via-ohci1394.rst for more 3636 info. 3637 3638 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands 3639 Rather than timing out after 20 ms if an EC 3640 command is not properly ACKed, override the length 3641 of the timeout. We have interrupts disabled while 3642 waiting for the ACK, so if this is set too high 3643 interrupts *may* be lost! 3644 3645 omap_mux= [OMAP] Override bootloader pin multiplexing. 3646 Format: <mux_mode0.mode_name=value>... 3647 For example, to override I2C bus2: 3648 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100 3649 3650 oprofile.timer= [HW] 3651 Use timer interrupt instead of performance counters 3652 3653 oprofile.cpu_type= Force an oprofile cpu type 3654 This might be useful if you have an older oprofile 3655 userland or if you want common events. 3656 Format: { arch_perfmon } 3657 arch_perfmon: [X86] Force use of architectural 3658 perfmon on Intel CPUs instead of the 3659 CPU specific event set. 3660 timer: [X86] Force use of architectural NMI 3661 timer mode (see also oprofile.timer 3662 for generic hr timer mode) 3663 3664 oops=panic Always panic on oopses. Default is to just kill the 3665 process, but there is a small probability of 3666 deadlocking the machine. 3667 This will also cause panics on machine check exceptions. 3668 Useful together with panic=30 to trigger a reboot. 3669 3670 page_alloc.shuffle= 3671 [KNL] Boolean flag to control whether the page allocator 3672 should randomize its free lists. The randomization may 3673 be automatically enabled if the kernel detects it is 3674 running on a platform with a direct-mapped memory-side 3675 cache, and this parameter can be used to 3676 override/disable that behavior. The state of the flag 3677 can be read from sysfs at: 3678 /sys/module/page_alloc/parameters/shuffle. 3679 3680 page_owner= [KNL] Boot-time page_owner enabling option. 3681 Storage of the information about who allocated 3682 each page is disabled in default. With this switch, 3683 we can turn it on. 3684 on: enable the feature 3685 3686 page_poison= [KNL] Boot-time parameter changing the state of 3687 poisoning on the buddy allocator, available with 3688 CONFIG_PAGE_POISONING=y. 3689 off: turn off poisoning (default) 3690 on: turn on poisoning 3691 3692 panic= [KNL] Kernel behaviour on panic: delay <timeout> 3693 timeout > 0: seconds before rebooting 3694 timeout = 0: wait forever 3695 timeout < 0: reboot immediately 3696 Format: <timeout> 3697 3698 panic_print= Bitmask for printing system info when panic happens. 3699 User can chose combination of the following bits: 3700 bit 0: print all tasks info 3701 bit 1: print system memory info 3702 bit 2: print timer info 3703 bit 3: print locks info if CONFIG_LOCKDEP is on 3704 bit 4: print ftrace buffer 3705 bit 5: print all printk messages in buffer 3706 3707 panic_on_taint= Bitmask for conditionally calling panic() in add_taint() 3708 Format: <hex>[,nousertaint] 3709 Hexadecimal bitmask representing the set of TAINT flags 3710 that will cause the kernel to panic when add_taint() is 3711 called with any of the flags in this set. 3712 The optional switch "nousertaint" can be utilized to 3713 prevent userspace forced crashes by writing to sysctl 3714 /proc/sys/kernel/tainted any flagset matching with the 3715 bitmask set on panic_on_taint. 3716 See Documentation/admin-guide/tainted-kernels.rst for 3717 extra details on the taint flags that users can pick 3718 to compose the bitmask to assign to panic_on_taint. 3719 3720 panic_on_warn panic() instead of WARN(). Useful to cause kdump 3721 on a WARN(). 3722 3723 pelt= [KNL] Boot-time parameter changing the PELT half life in ms 3724 Format: <int> 3725 32: Set the half life to 32ms 3726 8: Set the half life to 8ms 3727 default: 32 3728 3729 crash_kexec_post_notifiers 3730 Run kdump after running panic-notifiers and dumping 3731 kmsg. This only for the users who doubt kdump always 3732 succeeds in any situation. 3733 Note that this also increases risks of kdump failure, 3734 because some panic notifiers can make the crashed 3735 kernel more unstable. 3736 3737 parkbd.port= [HW] Parallel port number the keyboard adapter is 3738 connected to, default is 0. 3739 Format: <parport#> 3740 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation, 3741 0 for XT, 1 for AT (default is AT). 3742 Format: <mode> 3743 3744 parport= [HW,PPT] Specify parallel ports. 0 disables. 3745 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] } 3746 Use 'auto' to force the driver to use any 3747 IRQ/DMA settings detected (the default is to 3748 ignore detected IRQ/DMA settings because of 3749 possible conflicts). You can specify the base 3750 address, IRQ, and DMA settings; IRQ and DMA 3751 should be numbers, or 'auto' (for using detected 3752 settings on that particular port), or 'nofifo' 3753 (to avoid using a FIFO even if it is detected). 3754 Parallel ports are assigned in the order they 3755 are specified on the command line, starting 3756 with parport0. 3757 3758 parport_init_mode= [HW,PPT] 3759 Configure VIA parallel port to operate in 3760 a specific mode. This is necessary on Pegasos 3761 computer where firmware has no options for setting 3762 up parallel port mode and sets it to spp. 3763 Currently this function knows 686a and 8231 chips. 3764 Format: [spp|ps2|epp|ecp|ecpepp] 3765 3766 pause_on_oops= 3767 Halt all CPUs after the first oops has been printed for 3768 the specified number of seconds. This is to be used if 3769 your oopses keep scrolling off the screen. 3770 3771 pcbit= [HW,ISDN] 3772 3773 pcd. [PARIDE] 3774 See header of drivers/block/paride/pcd.c. 3775 See also Documentation/admin-guide/blockdev/paride.rst. 3776 3777 pci=option[,option...] [PCI] various PCI subsystem options. 3778 3779 Some options herein operate on a specific device 3780 or a set of devices (<pci_dev>). These are 3781 specified in one of the following formats: 3782 3783 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]* 3784 pci:<vendor>:<device>[:<subvendor>:<subdevice>] 3785 3786 Note: the first format specifies a PCI 3787 bus/device/function address which may change 3788 if new hardware is inserted, if motherboard 3789 firmware changes, or due to changes caused 3790 by other kernel parameters. If the 3791 domain is left unspecified, it is 3792 taken to be zero. Optionally, a path 3793 to a device through multiple device/function 3794 addresses can be specified after the base 3795 address (this is more robust against 3796 renumbering issues). The second format 3797 selects devices using IDs from the 3798 configuration space which may match multiple 3799 devices in the system. 3800 3801 earlydump dump PCI config space before the kernel 3802 changes anything 3803 off [X86] don't probe for the PCI bus 3804 bios [X86-32] force use of PCI BIOS, don't access 3805 the hardware directly. Use this if your machine 3806 has a non-standard PCI host bridge. 3807 nobios [X86-32] disallow use of PCI BIOS, only direct 3808 hardware access methods are allowed. Use this 3809 if you experience crashes upon bootup and you 3810 suspect they are caused by the BIOS. 3811 conf1 [X86] Force use of PCI Configuration Access 3812 Mechanism 1 (config address in IO port 0xCF8, 3813 data in IO port 0xCFC, both 32-bit). 3814 conf2 [X86] Force use of PCI Configuration Access 3815 Mechanism 2 (IO port 0xCF8 is an 8-bit port for 3816 the function, IO port 0xCFA, also 8-bit, sets 3817 bus number. The config space is then accessed 3818 through ports 0xC000-0xCFFF). 3819 See http://wiki.osdev.org/PCI for more info 3820 on the configuration access mechanisms. 3821 noaer [PCIE] If the PCIEAER kernel config parameter is 3822 enabled, this kernel boot option can be used to 3823 disable the use of PCIE advanced error reporting. 3824 nodomains [PCI] Disable support for multiple PCI 3825 root domains (aka PCI segments, in ACPI-speak). 3826 nommconf [X86] Disable use of MMCONFIG for PCI 3827 Configuration 3828 check_enable_amd_mmconf [X86] check for and enable 3829 properly configured MMIO access to PCI 3830 config space on AMD family 10h CPU 3831 nomsi [MSI] If the PCI_MSI kernel config parameter is 3832 enabled, this kernel boot option can be used to 3833 disable the use of MSI interrupts system-wide. 3834 noioapicquirk [APIC] Disable all boot interrupt quirks. 3835 Safety option to keep boot IRQs enabled. This 3836 should never be necessary. 3837 ioapicreroute [APIC] Enable rerouting of boot IRQs to the 3838 primary IO-APIC for bridges that cannot disable 3839 boot IRQs. This fixes a source of spurious IRQs 3840 when the system masks IRQs. 3841 noioapicreroute [APIC] Disable workaround that uses the 3842 boot IRQ equivalent of an IRQ that connects to 3843 a chipset where boot IRQs cannot be disabled. 3844 The opposite of ioapicreroute. 3845 biosirq [X86-32] Use PCI BIOS calls to get the interrupt 3846 routing table. These calls are known to be buggy 3847 on several machines and they hang the machine 3848 when used, but on other computers it's the only 3849 way to get the interrupt routing table. Try 3850 this option if the kernel is unable to allocate 3851 IRQs or discover secondary PCI buses on your 3852 motherboard. 3853 rom [X86] Assign address space to expansion ROMs. 3854 Use with caution as certain devices share 3855 address decoders between ROMs and other 3856 resources. 3857 norom [X86] Do not assign address space to 3858 expansion ROMs that do not already have 3859 BIOS assigned address ranges. 3860 nobar [X86] Do not assign address space to the 3861 BARs that weren't assigned by the BIOS. 3862 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be 3863 assigned automatically to PCI devices. You can 3864 make the kernel exclude IRQs of your ISA cards 3865 this way. 3866 pirqaddr=0xAAAAA [X86] Specify the physical address 3867 of the PIRQ table (normally generated 3868 by the BIOS) if it is outside the 3869 F0000h-100000h range. 3870 lastbus=N [X86] Scan all buses thru bus #N. Can be 3871 useful if the kernel is unable to find your 3872 secondary buses and you want to tell it 3873 explicitly which ones they are. 3874 assign-busses [X86] Always assign all PCI bus 3875 numbers ourselves, overriding 3876 whatever the firmware may have done. 3877 usepirqmask [X86] Honor the possible IRQ mask stored 3878 in the BIOS $PIR table. This is needed on 3879 some systems with broken BIOSes, notably 3880 some HP Pavilion N5400 and Omnibook XE3 3881 notebooks. This will have no effect if ACPI 3882 IRQ routing is enabled. 3883 noacpi [X86] Do not use ACPI for IRQ routing 3884 or for PCI scanning. 3885 use_crs [X86] Use PCI host bridge window information 3886 from ACPI. On BIOSes from 2008 or later, this 3887 is enabled by default. If you need to use this, 3888 please report a bug. 3889 nocrs [X86] Ignore PCI host bridge windows from ACPI. 3890 If you need to use this, please report a bug. 3891 routeirq Do IRQ routing for all PCI devices. 3892 This is normally done in pci_enable_device(), 3893 so this option is a temporary workaround 3894 for broken drivers that don't call it. 3895 skip_isa_align [X86] do not align io start addr, so can 3896 handle more pci cards 3897 noearly [X86] Don't do any early type 1 scanning. 3898 This might help on some broken boards which 3899 machine check when some devices' config space 3900 is read. But various workarounds are disabled 3901 and some IOMMU drivers will not work. 3902 bfsort Sort PCI devices into breadth-first order. 3903 This sorting is done to get a device 3904 order compatible with older (<= 2.4) kernels. 3905 nobfsort Don't sort PCI devices into breadth-first order. 3906 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size) 3907 tuning and use the BIOS-configured MPS defaults. 3908 pcie_bus_safe Set every device's MPS to the largest value 3909 supported by all devices below the root complex. 3910 pcie_bus_perf Set device MPS to the largest allowable MPS 3911 based on its parent bus. Also set MRRS (Max 3912 Read Request Size) to the largest supported 3913 value (no larger than the MPS that the device 3914 or bus can support) for best performance. 3915 pcie_bus_peer2peer Set every device's MPS to 128B, which 3916 every device is guaranteed to support. This 3917 configuration allows peer-to-peer DMA between 3918 any pair of devices, possibly at the cost of 3919 reduced performance. This also guarantees 3920 that hot-added devices will work. 3921 cbiosize=nn[KMG] The fixed amount of bus space which is 3922 reserved for the CardBus bridge's IO window. 3923 The default value is 256 bytes. 3924 cbmemsize=nn[KMG] The fixed amount of bus space which is 3925 reserved for the CardBus bridge's memory 3926 window. The default value is 64 megabytes. 3927 resource_alignment= 3928 Format: 3929 [<order of align>@]<pci_dev>[; ...] 3930 Specifies alignment and device to reassign 3931 aligned memory resources. How to 3932 specify the device is described above. 3933 If <order of align> is not specified, 3934 PAGE_SIZE is used as alignment. 3935 A PCI-PCI bridge can be specified if resource 3936 windows need to be expanded. 3937 To specify the alignment for several 3938 instances of a device, the PCI vendor, 3939 device, subvendor, and subdevice may be 3940 specified, e.g., 12@pci:8086:9c22:103c:198f 3941 for 4096-byte alignment. 3942 ecrc= Enable/disable PCIe ECRC (transaction layer 3943 end-to-end CRC checking). 3944 bios: Use BIOS/firmware settings. This is the 3945 the default. 3946 off: Turn ECRC off 3947 on: Turn ECRC on. 3948 hpiosize=nn[KMG] The fixed amount of bus space which is 3949 reserved for hotplug bridge's IO window. 3950 Default size is 256 bytes. 3951 hpmmiosize=nn[KMG] The fixed amount of bus space which is 3952 reserved for hotplug bridge's MMIO window. 3953 Default size is 2 megabytes. 3954 hpmmioprefsize=nn[KMG] The fixed amount of bus space which is 3955 reserved for hotplug bridge's MMIO_PREF window. 3956 Default size is 2 megabytes. 3957 hpmemsize=nn[KMG] The fixed amount of bus space which is 3958 reserved for hotplug bridge's MMIO and 3959 MMIO_PREF window. 3960 Default size is 2 megabytes. 3961 hpbussize=nn The minimum amount of additional bus numbers 3962 reserved for buses below a hotplug bridge. 3963 Default is 1. 3964 realloc= Enable/disable reallocating PCI bridge resources 3965 if allocations done by BIOS are too small to 3966 accommodate resources required by all child 3967 devices. 3968 off: Turn realloc off 3969 on: Turn realloc on 3970 realloc same as realloc=on 3971 noari do not use PCIe ARI. 3972 noats [PCIE, Intel-IOMMU, AMD-IOMMU] 3973 do not use PCIe ATS (and IOMMU device IOTLB). 3974 pcie_scan_all Scan all possible PCIe devices. Otherwise we 3975 only look for one device below a PCIe downstream 3976 port. 3977 big_root_window Try to add a big 64bit memory window to the PCIe 3978 root complex on AMD CPUs. Some GFX hardware 3979 can resize a BAR to allow access to all VRAM. 3980 Adding the window is slightly risky (it may 3981 conflict with unreported devices), so this 3982 taints the kernel. 3983 disable_acs_redir=<pci_dev>[; ...] 3984 Specify one or more PCI devices (in the format 3985 specified above) separated by semicolons. 3986 Each device specified will have the PCI ACS 3987 redirect capabilities forced off which will 3988 allow P2P traffic between devices through 3989 bridges without forcing it upstream. Note: 3990 this removes isolation between devices and 3991 may put more devices in an IOMMU group. 3992 force_floating [S390] Force usage of floating interrupts. 3993 nomio [S390] Do not use MIO instructions. 3994 norid [S390] ignore the RID field and force use of 3995 one PCI domain per PCI function 3996 3997 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power 3998 Management. 3999 off Disable ASPM. 4000 force Enable ASPM even on devices that claim not to support it. 4001 WARNING: Forcing ASPM on may cause system lockups. 4002 4003 pcie_ports= [PCIE] PCIe port services handling: 4004 native Use native PCIe services (PME, AER, DPC, PCIe hotplug) 4005 even if the platform doesn't give the OS permission to 4006 use them. This may cause conflicts if the platform 4007 also tries to use these services. 4008 dpc-native Use native PCIe service for DPC only. May 4009 cause conflicts if firmware uses AER or DPC. 4010 compat Disable native PCIe services (PME, AER, DPC, PCIe 4011 hotplug). 4012 4013 pcie_port_pm= [PCIE] PCIe port power management handling: 4014 off Disable power management of all PCIe ports 4015 force Forcibly enable power management of all PCIe ports 4016 4017 pcie_pme= [PCIE,PM] Native PCIe PME signaling options: 4018 nomsi Do not use MSI for native PCIe PME signaling (this makes 4019 all PCIe root ports use INTx for all services). 4020 4021 pcmv= [HW,PCMCIA] BadgePAD 4 4022 4023 pd_ignore_unused 4024 [PM] 4025 Keep all power-domains already enabled by bootloader on, 4026 even if no driver has claimed them. This is useful 4027 for debug and development, but should not be 4028 needed on a platform with proper driver support. 4029 4030 pd. [PARIDE] 4031 See Documentation/admin-guide/blockdev/paride.rst. 4032 4033 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at 4034 boot time. 4035 Format: { 0 | 1 } 4036 See arch/parisc/kernel/pdc_chassis.c 4037 4038 percpu_alloc= Select which percpu first chunk allocator to use. 4039 Currently supported values are "embed" and "page". 4040 Archs may support subset or none of the selections. 4041 See comments in mm/percpu.c for details on each 4042 allocator. This parameter is primarily for debugging 4043 and performance comparison. 4044 4045 pf. [PARIDE] 4046 See Documentation/admin-guide/blockdev/paride.rst. 4047 4048 pg. [PARIDE] 4049 See Documentation/admin-guide/blockdev/paride.rst. 4050 4051 pirq= [SMP,APIC] Manual mp-table setup 4052 See Documentation/x86/i386/IO-APIC.rst. 4053 4054 plip= [PPT,NET] Parallel port network link 4055 Format: { parport<nr> | timid | 0 } 4056 See also Documentation/admin-guide/parport.rst. 4057 4058 pmtmr= [X86] Manual setup of pmtmr I/O Port. 4059 Override pmtimer IOPort with a hex value. 4060 e.g. pmtmr=0x508 4061 4062 pm_debug_messages [SUSPEND,KNL] 4063 Enable suspend/resume debug messages during boot up. 4064 4065 pnp.debug=1 [PNP] 4066 Enable PNP debug messages (depends on the 4067 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time 4068 via /sys/module/pnp/parameters/debug. We always show 4069 current resource usage; turning this on also shows 4070 possible settings and some assignment information. 4071 4072 pnpacpi= [ACPI] 4073 { off } 4074 4075 pnpbios= [ISAPNP] 4076 { on | off | curr | res | no-curr | no-res } 4077 4078 pnp_reserve_irq= 4079 [ISAPNP] Exclude IRQs for the autoconfiguration 4080 4081 pnp_reserve_dma= 4082 [ISAPNP] Exclude DMAs for the autoconfiguration 4083 4084 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration 4085 Ranges are in pairs (I/O port base and size). 4086 4087 pnp_reserve_mem= 4088 [ISAPNP] Exclude memory regions for the 4089 autoconfiguration. 4090 Ranges are in pairs (memory base and size). 4091 4092 ports= [IP_VS_FTP] IPVS ftp helper module 4093 Default is 21. 4094 Up to 8 (IP_VS_APP_MAX_PORTS) ports 4095 may be specified. 4096 Format: <port>,<port>.... 4097 4098 powersave=off [PPC] This option disables power saving features. 4099 It specifically disables cpuidle and sets the 4100 platform machine description specific power_save 4101 function to NULL. On Idle the CPU just reduces 4102 execution priority. 4103 4104 ppc_strict_facility_enable 4105 [PPC] This option catches any kernel floating point, 4106 Altivec, VSX and SPE outside of regions specifically 4107 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()). 4108 There is some performance impact when enabling this. 4109 4110 ppc_tm= [PPC] 4111 Format: {"off"} 4112 Disable Hardware Transactional Memory 4113 4114 print-fatal-signals= 4115 [KNL] debug: print fatal signals 4116 4117 If enabled, warn about various signal handling 4118 related application anomalies: too many signals, 4119 too many POSIX.1 timers, fatal signals causing a 4120 coredump - etc. 4121 4122 If you hit the warning due to signal overflow, 4123 you might want to try "ulimit -i unlimited". 4124 4125 default: off. 4126 4127 printk.always_kmsg_dump= 4128 Trigger kmsg_dump for cases other than kernel oops or 4129 panics 4130 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 4131 default: disabled 4132 4133 printk.devkmsg={on,off,ratelimit} 4134 Control writing to /dev/kmsg. 4135 on - unlimited logging to /dev/kmsg from userspace 4136 off - logging to /dev/kmsg disabled 4137 ratelimit - ratelimit the logging 4138 Default: ratelimit 4139 4140 printk.time= Show timing data prefixed to each printk message line 4141 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 4142 4143 processor.max_cstate= [HW,ACPI] 4144 Limit processor to maximum C-state 4145 max_cstate=9 overrides any DMI blacklist limit. 4146 4147 processor.nocst [HW,ACPI] 4148 Ignore the _CST method to determine C-states, 4149 instead using the legacy FADT method 4150 4151 profile= [KNL] Enable kernel profiling via /proc/profile 4152 Format: [<profiletype>,]<number> 4153 Param: <profiletype>: "schedule", "sleep", or "kvm" 4154 [defaults to kernel profiling] 4155 Param: "schedule" - profile schedule points. 4156 Param: "sleep" - profile D-state sleeping (millisecs). 4157 Requires CONFIG_SCHEDSTATS 4158 Param: "kvm" - profile VM exits. 4159 Param: <number> - step/bucket size as a power of 2 for 4160 statistical time based profiling. 4161 4162 prompt_ramdisk= [RAM] [Deprecated] 4163 4164 prot_virt= [S390] enable hosting protected virtual machines 4165 isolated from the hypervisor (if hardware supports 4166 that). 4167 Format: <bool> 4168 4169 psi= [KNL] Enable or disable pressure stall information 4170 tracking. 4171 Format: <bool> 4172 4173 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to 4174 probe for; one of (bare|imps|exps|lifebook|any). 4175 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports 4176 per second. 4177 psmouse.resetafter= [HW,MOUSE] 4178 Try to reset the device after so many bad packets 4179 (0 = never). 4180 psmouse.resolution= 4181 [HW,MOUSE] Set desired mouse resolution, in dpi. 4182 psmouse.smartscroll= 4183 [HW,MOUSE] Controls Logitech smartscroll autorepeat. 4184 0 = disabled, 1 = enabled (default). 4185 4186 pstore.backend= Specify the name of the pstore backend to use 4187 4188 pt. [PARIDE] 4189 See Documentation/admin-guide/blockdev/paride.rst. 4190 4191 pti= [X86-64] Control Page Table Isolation of user and 4192 kernel address spaces. Disabling this feature 4193 removes hardening, but improves performance of 4194 system calls and interrupts. 4195 4196 on - unconditionally enable 4197 off - unconditionally disable 4198 auto - kernel detects whether your CPU model is 4199 vulnerable to issues that PTI mitigates 4200 4201 Not specifying this option is equivalent to pti=auto. 4202 4203 nopti [X86-64] 4204 Equivalent to pti=off 4205 4206 pty.legacy_count= 4207 [KNL] Number of legacy pty's. Overwrites compiled-in 4208 default number. 4209 4210 quiet [KNL] Disable most log messages 4211 4212 r128= [HW,DRM] 4213 4214 raid= [HW,RAID] 4215 See Documentation/admin-guide/md.rst. 4216 4217 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes 4218 See Documentation/admin-guide/blockdev/ramdisk.rst. 4219 4220 ramdisk_start= [RAM] RAM disk image start address 4221 4222 random.trust_cpu={on,off} 4223 [KNL] Enable or disable trusting the use of the 4224 CPU's random number generator (if available) to 4225 fully seed the kernel's CRNG. Default is controlled 4226 by CONFIG_RANDOM_TRUST_CPU. 4227 4228 random.trust_bootloader={on,off} 4229 [KNL] Enable or disable trusting the use of a 4230 seed passed by the bootloader (if available) to 4231 fully seed the kernel's CRNG. Default is controlled 4232 by CONFIG_RANDOM_TRUST_BOOTLOADER. 4233 4234 ras=option[,option,...] [KNL] RAS-specific options 4235 4236 cec_disable [X86] 4237 Disable the Correctable Errors Collector, 4238 see CONFIG_RAS_CEC help text. 4239 4240 rcu_nocbs= [KNL] 4241 The argument is a cpu list, as described above, 4242 except that the string "all" can be used to 4243 specify every CPU on the system. 4244 4245 In kernels built with CONFIG_RCU_NOCB_CPU=y, set 4246 the specified list of CPUs to be no-callback CPUs. 4247 Invocation of these CPUs' RCU callbacks will be 4248 offloaded to "rcuox/N" kthreads created for that 4249 purpose, where "x" is "p" for RCU-preempt, and 4250 "s" for RCU-sched, and "N" is the CPU number. 4251 This reduces OS jitter on the offloaded CPUs, 4252 which can be useful for HPC and real-time 4253 workloads. It can also improve energy efficiency 4254 for asymmetric multiprocessors. 4255 4256 rcu_nocb_poll [KNL] 4257 Rather than requiring that offloaded CPUs 4258 (specified by rcu_nocbs= above) explicitly 4259 awaken the corresponding "rcuoN" kthreads, 4260 make these kthreads poll for callbacks. 4261 This improves the real-time response for the 4262 offloaded CPUs by relieving them of the need to 4263 wake up the corresponding kthread, but degrades 4264 energy efficiency by requiring that the kthreads 4265 periodically wake up to do the polling. 4266 4267 rcutree.blimit= [KNL] 4268 Set maximum number of finished RCU callbacks to 4269 process in one batch. 4270 4271 rcutree.dump_tree= [KNL] 4272 Dump the structure of the rcu_node combining tree 4273 out at early boot. This is used for diagnostic 4274 purposes, to verify correct tree setup. 4275 4276 rcutree.gp_cleanup_delay= [KNL] 4277 Set the number of jiffies to delay each step of 4278 RCU grace-period cleanup. 4279 4280 rcutree.gp_init_delay= [KNL] 4281 Set the number of jiffies to delay each step of 4282 RCU grace-period initialization. 4283 4284 rcutree.gp_preinit_delay= [KNL] 4285 Set the number of jiffies to delay each step of 4286 RCU grace-period pre-initialization, that is, 4287 the propagation of recent CPU-hotplug changes up 4288 the rcu_node combining tree. 4289 4290 rcutree.use_softirq= [KNL] 4291 If set to zero, move all RCU_SOFTIRQ processing to 4292 per-CPU rcuc kthreads. Defaults to a non-zero 4293 value, meaning that RCU_SOFTIRQ is used by default. 4294 Specify rcutree.use_softirq=0 to use rcuc kthreads. 4295 4296 rcutree.rcu_fanout_exact= [KNL] 4297 Disable autobalancing of the rcu_node combining 4298 tree. This is used by rcutorture, and might 4299 possibly be useful for architectures having high 4300 cache-to-cache transfer latencies. 4301 4302 rcutree.rcu_fanout_leaf= [KNL] 4303 Change the number of CPUs assigned to each 4304 leaf rcu_node structure. Useful for very 4305 large systems, which will choose the value 64, 4306 and for NUMA systems with large remote-access 4307 latencies, which will choose a value aligned 4308 with the appropriate hardware boundaries. 4309 4310 rcutree.rcu_min_cached_objs= [KNL] 4311 Minimum number of objects which are cached and 4312 maintained per one CPU. Object size is equal 4313 to PAGE_SIZE. The cache allows to reduce the 4314 pressure to page allocator, also it makes the 4315 whole algorithm to behave better in low memory 4316 condition. 4317 4318 rcutree.jiffies_till_first_fqs= [KNL] 4319 Set delay from grace-period initialization to 4320 first attempt to force quiescent states. 4321 Units are jiffies, minimum value is zero, 4322 and maximum value is HZ. 4323 4324 rcutree.jiffies_till_next_fqs= [KNL] 4325 Set delay between subsequent attempts to force 4326 quiescent states. Units are jiffies, minimum 4327 value is one, and maximum value is HZ. 4328 4329 rcutree.jiffies_till_sched_qs= [KNL] 4330 Set required age in jiffies for a 4331 given grace period before RCU starts 4332 soliciting quiescent-state help from 4333 rcu_note_context_switch() and cond_resched(). 4334 If not specified, the kernel will calculate 4335 a value based on the most recent settings 4336 of rcutree.jiffies_till_first_fqs 4337 and rcutree.jiffies_till_next_fqs. 4338 This calculated value may be viewed in 4339 rcutree.jiffies_to_sched_qs. Any attempt to set 4340 rcutree.jiffies_to_sched_qs will be cheerfully 4341 overwritten. 4342 4343 rcutree.kthread_prio= [KNL,BOOT] 4344 Set the SCHED_FIFO priority of the RCU per-CPU 4345 kthreads (rcuc/N). This value is also used for 4346 the priority of the RCU boost threads (rcub/N) 4347 and for the RCU grace-period kthreads (rcu_bh, 4348 rcu_preempt, and rcu_sched). If RCU_BOOST is 4349 set, valid values are 1-99 and the default is 1 4350 (the least-favored priority). Otherwise, when 4351 RCU_BOOST is not set, valid values are 0-99 and 4352 the default is zero (non-realtime operation). 4353 4354 rcutree.rcu_nocb_gp_stride= [KNL] 4355 Set the number of NOCB callback kthreads in 4356 each group, which defaults to the square root 4357 of the number of CPUs. Larger numbers reduce 4358 the wakeup overhead on the global grace-period 4359 kthread, but increases that same overhead on 4360 each group's NOCB grace-period kthread. 4361 4362 rcutree.qhimark= [KNL] 4363 Set threshold of queued RCU callbacks beyond which 4364 batch limiting is disabled. 4365 4366 rcutree.qlowmark= [KNL] 4367 Set threshold of queued RCU callbacks below which 4368 batch limiting is re-enabled. 4369 4370 rcutree.qovld= [KNL] 4371 Set threshold of queued RCU callbacks beyond which 4372 RCU's force-quiescent-state scan will aggressively 4373 enlist help from cond_resched() and sched IPIs to 4374 help CPUs more quickly reach quiescent states. 4375 Set to less than zero to make this be set based 4376 on rcutree.qhimark at boot time and to zero to 4377 disable more aggressive help enlistment. 4378 4379 rcutree.rcu_idle_gp_delay= [KNL] 4380 Set wakeup interval for idle CPUs that have 4381 RCU callbacks (RCU_FAST_NO_HZ=y). 4382 4383 rcutree.rcu_idle_lazy_gp_delay= [KNL] 4384 Set wakeup interval for idle CPUs that have 4385 only "lazy" RCU callbacks (RCU_FAST_NO_HZ=y). 4386 Lazy RCU callbacks are those which RCU can 4387 prove do nothing more than free memory. 4388 4389 rcutree.rcu_kick_kthreads= [KNL] 4390 Cause the grace-period kthread to get an extra 4391 wake_up() if it sleeps three times longer than 4392 it should at force-quiescent-state time. 4393 This wake_up() will be accompanied by a 4394 WARN_ONCE() splat and an ftrace_dump(). 4395 4396 rcutree.rcu_unlock_delay= [KNL] 4397 In CONFIG_RCU_STRICT_GRACE_PERIOD=y kernels, 4398 this specifies an rcu_read_unlock()-time delay 4399 in microseconds. This defaults to zero. 4400 Larger delays increase the probability of 4401 catching RCU pointer leaks, that is, buggy use 4402 of RCU-protected pointers after the relevant 4403 rcu_read_unlock() has completed. 4404 4405 rcutree.sysrq_rcu= [KNL] 4406 Commandeer a sysrq key to dump out Tree RCU's 4407 rcu_node tree with an eye towards determining 4408 why a new grace period has not yet started. 4409 4410 rcuscale.gp_async= [KNL] 4411 Measure performance of asynchronous 4412 grace-period primitives such as call_rcu(). 4413 4414 rcuscale.gp_async_max= [KNL] 4415 Specify the maximum number of outstanding 4416 callbacks per writer thread. When a writer 4417 thread exceeds this limit, it invokes the 4418 corresponding flavor of rcu_barrier() to allow 4419 previously posted callbacks to drain. 4420 4421 rcuscale.gp_exp= [KNL] 4422 Measure performance of expedited synchronous 4423 grace-period primitives. 4424 4425 rcuscale.holdoff= [KNL] 4426 Set test-start holdoff period. The purpose of 4427 this parameter is to delay the start of the 4428 test until boot completes in order to avoid 4429 interference. 4430 4431 rcuscale.kfree_rcu_test= [KNL] 4432 Set to measure performance of kfree_rcu() flooding. 4433 4434 rcuscale.kfree_nthreads= [KNL] 4435 The number of threads running loops of kfree_rcu(). 4436 4437 rcuscale.kfree_alloc_num= [KNL] 4438 Number of allocations and frees done in an iteration. 4439 4440 rcuscale.kfree_loops= [KNL] 4441 Number of loops doing rcuscale.kfree_alloc_num number 4442 of allocations and frees. 4443 4444 rcuscale.nreaders= [KNL] 4445 Set number of RCU readers. The value -1 selects 4446 N, where N is the number of CPUs. A value 4447 "n" less than -1 selects N-n+1, where N is again 4448 the number of CPUs. For example, -2 selects N 4449 (the number of CPUs), -3 selects N+1, and so on. 4450 A value of "n" less than or equal to -N selects 4451 a single reader. 4452 4453 rcuscale.nwriters= [KNL] 4454 Set number of RCU writers. The values operate 4455 the same as for rcuscale.nreaders. 4456 N, where N is the number of CPUs 4457 4458 rcuscale.perf_type= [KNL] 4459 Specify the RCU implementation to test. 4460 4461 rcuscale.shutdown= [KNL] 4462 Shut the system down after performance tests 4463 complete. This is useful for hands-off automated 4464 testing. 4465 4466 rcuscale.verbose= [KNL] 4467 Enable additional printk() statements. 4468 4469 rcuscale.writer_holdoff= [KNL] 4470 Write-side holdoff between grace periods, 4471 in microseconds. The default of zero says 4472 no holdoff. 4473 4474 rcutorture.fqs_duration= [KNL] 4475 Set duration of force_quiescent_state bursts 4476 in microseconds. 4477 4478 rcutorture.fqs_holdoff= [KNL] 4479 Set holdoff time within force_quiescent_state bursts 4480 in microseconds. 4481 4482 rcutorture.fqs_stutter= [KNL] 4483 Set wait time between force_quiescent_state bursts 4484 in seconds. 4485 4486 rcutorture.fwd_progress= [KNL] 4487 Enable RCU grace-period forward-progress testing 4488 for the types of RCU supporting this notion. 4489 4490 rcutorture.fwd_progress_div= [KNL] 4491 Specify the fraction of a CPU-stall-warning 4492 period to do tight-loop forward-progress testing. 4493 4494 rcutorture.fwd_progress_holdoff= [KNL] 4495 Number of seconds to wait between successive 4496 forward-progress tests. 4497 4498 rcutorture.fwd_progress_need_resched= [KNL] 4499 Enclose cond_resched() calls within checks for 4500 need_resched() during tight-loop forward-progress 4501 testing. 4502 4503 rcutorture.gp_cond= [KNL] 4504 Use conditional/asynchronous update-side 4505 primitives, if available. 4506 4507 rcutorture.gp_exp= [KNL] 4508 Use expedited update-side primitives, if available. 4509 4510 rcutorture.gp_normal= [KNL] 4511 Use normal (non-expedited) asynchronous 4512 update-side primitives, if available. 4513 4514 rcutorture.gp_sync= [KNL] 4515 Use normal (non-expedited) synchronous 4516 update-side primitives, if available. If all 4517 of rcutorture.gp_cond=, rcutorture.gp_exp=, 4518 rcutorture.gp_normal=, and rcutorture.gp_sync= 4519 are zero, rcutorture acts as if is interpreted 4520 they are all non-zero. 4521 4522 rcutorture.irqreader= [KNL] 4523 Run RCU readers from irq handlers, or, more 4524 accurately, from a timer handler. Not all RCU 4525 flavors take kindly to this sort of thing. 4526 4527 rcutorture.leakpointer= [KNL] 4528 Leak an RCU-protected pointer out of the reader. 4529 This can of course result in splats, and is 4530 intended to test the ability of things like 4531 CONFIG_RCU_STRICT_GRACE_PERIOD=y to detect 4532 such leaks. 4533 4534 rcutorture.n_barrier_cbs= [KNL] 4535 Set callbacks/threads for rcu_barrier() testing. 4536 4537 rcutorture.nfakewriters= [KNL] 4538 Set number of concurrent RCU writers. These just 4539 stress RCU, they don't participate in the actual 4540 test, hence the "fake". 4541 4542 rcutorture.nreaders= [KNL] 4543 Set number of RCU readers. The value -1 selects 4544 N-1, where N is the number of CPUs. A value 4545 "n" less than -1 selects N-n-2, where N is again 4546 the number of CPUs. For example, -2 selects N 4547 (the number of CPUs), -3 selects N+1, and so on. 4548 4549 rcutorture.object_debug= [KNL] 4550 Enable debug-object double-call_rcu() testing. 4551 4552 rcutorture.onoff_holdoff= [KNL] 4553 Set time (s) after boot for CPU-hotplug testing. 4554 4555 rcutorture.onoff_interval= [KNL] 4556 Set time (jiffies) between CPU-hotplug operations, 4557 or zero to disable CPU-hotplug testing. 4558 4559 rcutorture.read_exit= [KNL] 4560 Set the number of read-then-exit kthreads used 4561 to test the interaction of RCU updaters and 4562 task-exit processing. 4563 4564 rcutorture.read_exit_burst= [KNL] 4565 The number of times in a given read-then-exit 4566 episode that a set of read-then-exit kthreads 4567 is spawned. 4568 4569 rcutorture.read_exit_delay= [KNL] 4570 The delay, in seconds, between successive 4571 read-then-exit testing episodes. 4572 4573 rcutorture.shuffle_interval= [KNL] 4574 Set task-shuffle interval (s). Shuffling tasks 4575 allows some CPUs to go into dyntick-idle mode 4576 during the rcutorture test. 4577 4578 rcutorture.shutdown_secs= [KNL] 4579 Set time (s) after boot system shutdown. This 4580 is useful for hands-off automated testing. 4581 4582 rcutorture.stall_cpu= [KNL] 4583 Duration of CPU stall (s) to test RCU CPU stall 4584 warnings, zero to disable. 4585 4586 rcutorture.stall_cpu_block= [KNL] 4587 Sleep while stalling if set. This will result 4588 in warnings from preemptible RCU in addition 4589 to any other stall-related activity. 4590 4591 rcutorture.stall_cpu_holdoff= [KNL] 4592 Time to wait (s) after boot before inducing stall. 4593 4594 rcutorture.stall_cpu_irqsoff= [KNL] 4595 Disable interrupts while stalling if set. 4596 4597 rcutorture.stall_gp_kthread= [KNL] 4598 Duration (s) of forced sleep within RCU 4599 grace-period kthread to test RCU CPU stall 4600 warnings, zero to disable. If both stall_cpu 4601 and stall_gp_kthread are specified, the 4602 kthread is starved first, then the CPU. 4603 4604 rcutorture.stat_interval= [KNL] 4605 Time (s) between statistics printk()s. 4606 4607 rcutorture.stutter= [KNL] 4608 Time (s) to stutter testing, for example, specifying 4609 five seconds causes the test to run for five seconds, 4610 wait for five seconds, and so on. This tests RCU's 4611 ability to transition abruptly to and from idle. 4612 4613 rcutorture.test_boost= [KNL] 4614 Test RCU priority boosting? 0=no, 1=maybe, 2=yes. 4615 "Maybe" means test if the RCU implementation 4616 under test support RCU priority boosting. 4617 4618 rcutorture.test_boost_duration= [KNL] 4619 Duration (s) of each individual boost test. 4620 4621 rcutorture.test_boost_interval= [KNL] 4622 Interval (s) between each boost test. 4623 4624 rcutorture.test_no_idle_hz= [KNL] 4625 Test RCU's dyntick-idle handling. See also the 4626 rcutorture.shuffle_interval parameter. 4627 4628 rcutorture.torture_type= [KNL] 4629 Specify the RCU implementation to test. 4630 4631 rcutorture.verbose= [KNL] 4632 Enable additional printk() statements. 4633 4634 rcupdate.rcu_cpu_stall_ftrace_dump= [KNL] 4635 Dump ftrace buffer after reporting RCU CPU 4636 stall warning. 4637 4638 rcupdate.rcu_cpu_stall_suppress= [KNL] 4639 Suppress RCU CPU stall warning messages. 4640 4641 rcupdate.rcu_cpu_stall_suppress_at_boot= [KNL] 4642 Suppress RCU CPU stall warning messages and 4643 rcutorture writer stall warnings that occur 4644 during early boot, that is, during the time 4645 before the init task is spawned. 4646 4647 rcupdate.rcu_cpu_stall_timeout= [KNL] 4648 Set timeout for RCU CPU stall warning messages. 4649 4650 rcupdate.rcu_expedited= [KNL] 4651 Use expedited grace-period primitives, for 4652 example, synchronize_rcu_expedited() instead 4653 of synchronize_rcu(). This reduces latency, 4654 but can increase CPU utilization, degrade 4655 real-time latency, and degrade energy efficiency. 4656 No effect on CONFIG_TINY_RCU kernels. 4657 4658 rcupdate.rcu_normal= [KNL] 4659 Use only normal grace-period primitives, 4660 for example, synchronize_rcu() instead of 4661 synchronize_rcu_expedited(). This improves 4662 real-time latency, CPU utilization, and 4663 energy efficiency, but can expose users to 4664 increased grace-period latency. This parameter 4665 overrides rcupdate.rcu_expedited. No effect on 4666 CONFIG_TINY_RCU kernels. 4667 4668 rcupdate.rcu_normal_after_boot= [KNL] 4669 Once boot has completed (that is, after 4670 rcu_end_inkernel_boot() has been invoked), use 4671 only normal grace-period primitives. No effect 4672 on CONFIG_TINY_RCU kernels. 4673 4674 rcupdate.rcu_task_ipi_delay= [KNL] 4675 Set time in jiffies during which RCU tasks will 4676 avoid sending IPIs, starting with the beginning 4677 of a given grace period. Setting a large 4678 number avoids disturbing real-time workloads, 4679 but lengthens grace periods. 4680 4681 rcupdate.rcu_task_stall_timeout= [KNL] 4682 Set timeout in jiffies for RCU task stall warning 4683 messages. Disable with a value less than or equal 4684 to zero. 4685 4686 rcupdate.rcu_self_test= [KNL] 4687 Run the RCU early boot self tests 4688 4689 rdinit= [KNL] 4690 Format: <full_path> 4691 Run specified binary instead of /init from the ramdisk, 4692 used for early userspace startup. See initrd. 4693 4694 rdrand= [X86] 4695 force - Override the decision by the kernel to hide the 4696 advertisement of RDRAND support (this affects 4697 certain AMD processors because of buggy BIOS 4698 support, specifically around the suspend/resume 4699 path). 4700 4701 rdt= [HW,X86,RDT] 4702 Turn on/off individual RDT features. List is: 4703 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp, 4704 mba. 4705 E.g. to turn on cmt and turn off mba use: 4706 rdt=cmt,!mba 4707 4708 reboot= [KNL] 4709 Format (x86 or x86_64): 4710 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] \ 4711 [[,]s[mp]#### \ 4712 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \ 4713 [[,]f[orce] 4714 Where reboot_mode is one of warm (soft) or cold (hard) or gpio 4715 (prefix with 'panic_' to set mode for panic 4716 reboot only), 4717 reboot_type is one of bios, acpi, kbd, triple, efi, or pci, 4718 reboot_force is either force or not specified, 4719 reboot_cpu is s[mp]#### with #### being the processor 4720 to be used for rebooting. 4721 4722 refscale.holdoff= [KNL] 4723 Set test-start holdoff period. The purpose of 4724 this parameter is to delay the start of the 4725 test until boot completes in order to avoid 4726 interference. 4727 4728 refscale.loops= [KNL] 4729 Set the number of loops over the synchronization 4730 primitive under test. Increasing this number 4731 reduces noise due to loop start/end overhead, 4732 but the default has already reduced the per-pass 4733 noise to a handful of picoseconds on ca. 2020 4734 x86 laptops. 4735 4736 refscale.nreaders= [KNL] 4737 Set number of readers. The default value of -1 4738 selects N, where N is roughly 75% of the number 4739 of CPUs. A value of zero is an interesting choice. 4740 4741 refscale.nruns= [KNL] 4742 Set number of runs, each of which is dumped onto 4743 the console log. 4744 4745 refscale.readdelay= [KNL] 4746 Set the read-side critical-section duration, 4747 measured in microseconds. 4748 4749 refscale.scale_type= [KNL] 4750 Specify the read-protection implementation to test. 4751 4752 refscale.shutdown= [KNL] 4753 Shut down the system at the end of the performance 4754 test. This defaults to 1 (shut it down) when 4755 refscale is built into the kernel and to 0 (leave 4756 it running) when refscale is built as a module. 4757 4758 refscale.verbose= [KNL] 4759 Enable additional printk() statements. 4760 4761 relax_domain_level= 4762 [KNL, SMP] Set scheduler's default relax_domain_level. 4763 See Documentation/admin-guide/cgroup-v1/cpusets.rst. 4764 4765 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory 4766 Format: <base1>,<size1>[,<base2>,<size2>,...] 4767 Reserve I/O ports or memory so the kernel won't use 4768 them. If <base> is less than 0x10000, the region 4769 is assumed to be I/O ports; otherwise it is memory. 4770 4771 reservetop= [X86-32] 4772 Format: nn[KMG] 4773 Reserves a hole at the top of the kernel virtual 4774 address space. 4775 4776 reservelow= [X86] 4777 Format: nn[K] 4778 Set the amount of memory to reserve for BIOS at 4779 the bottom of the address space. 4780 4781 reset_devices [KNL] Force drivers to reset the underlying device 4782 during initialization. 4783 4784 resume= [SWSUSP] 4785 Specify the partition device for software suspend 4786 Format: 4787 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>} 4788 4789 resume_offset= [SWSUSP] 4790 Specify the offset from the beginning of the partition 4791 given by "resume=" at which the swap header is located, 4792 in <PAGE_SIZE> units (needed only for swap files). 4793 See Documentation/power/swsusp-and-swap-files.rst 4794 4795 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to 4796 read the resume files 4797 4798 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up. 4799 Useful for devices that are detected asynchronously 4800 (e.g. USB and MMC devices). 4801 4802 hibernate= [HIBERNATION] 4803 noresume Don't check if there's a hibernation image 4804 present during boot. 4805 nocompress Don't compress/decompress hibernation images. 4806 no Disable hibernation and resume. 4807 protect_image Turn on image protection during restoration 4808 (that will set all pages holding image data 4809 during restoration read-only). 4810 4811 retain_initrd [RAM] Keep initrd memory after extraction 4812 4813 retbleed= [X86] Control mitigation of RETBleed (Arbitrary 4814 Speculative Code Execution with Return Instructions) 4815 vulnerability. 4816 4817 AMD-based UNRET and IBPB mitigations alone do not stop 4818 sibling threads from influencing the predictions of other 4819 sibling threads. For that reason, STIBP is used on pro- 4820 cessors that support it, and mitigate SMT on processors 4821 that don't. 4822 4823 off - no mitigation 4824 auto - automatically select a migitation 4825 auto,nosmt - automatically select a mitigation, 4826 disabling SMT if necessary for 4827 the full mitigation (only on Zen1 4828 and older without STIBP). 4829 ibpb - On AMD, mitigate short speculation 4830 windows on basic block boundaries too. 4831 Safe, highest perf impact. It also 4832 enables STIBP if present. Not suitable 4833 on Intel. 4834 ibpb,nosmt - Like "ibpb" above but will disable SMT 4835 when STIBP is not available. This is 4836 the alternative for systems which do not 4837 have STIBP. 4838 unret - Force enable untrained return thunks, 4839 only effective on AMD f15h-f17h based 4840 systems. 4841 unret,nosmt - Like unret, but will disable SMT when STIBP 4842 is not available. This is the alternative for 4843 systems which do not have STIBP. 4844 4845 Selecting 'auto' will choose a mitigation method at run 4846 time according to the CPU. 4847 4848 Not specifying this option is equivalent to retbleed=auto. 4849 4850 rfkill.default_state= 4851 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm, 4852 etc. communication is blocked by default. 4853 1 Unblocked. 4854 4855 rfkill.master_switch_mode= 4856 0 The "airplane mode" button does nothing. 4857 1 The "airplane mode" button toggles between everything 4858 blocked and the previous configuration. 4859 2 The "airplane mode" button toggles between everything 4860 blocked and everything unblocked. 4861 4862 rhash_entries= [KNL,NET] 4863 Set number of hash buckets for route cache 4864 4865 ring3mwait=disable 4866 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported 4867 CPUs. 4868 4869 ro [KNL] Mount root device read-only on boot 4870 4871 rodata= [KNL] 4872 on Mark read-only kernel memory as read-only (default). 4873 off Leave read-only kernel memory writable for debugging. 4874 4875 rockchip.usb_uart 4876 Enable the uart passthrough on the designated usb port 4877 on Rockchip SoCs. When active, the signals of the 4878 debug-uart get routed to the D+ and D- pins of the usb 4879 port and the regular usb controller gets disabled. 4880 4881 root= [KNL] Root filesystem 4882 See name_to_dev_t comment in init/do_mounts.c. 4883 4884 rootdelay= [KNL] Delay (in seconds) to pause before attempting to 4885 mount the root filesystem 4886 4887 rootflags= [KNL] Set root filesystem mount option string 4888 4889 rootfstype= [KNL] Set root filesystem type 4890 4891 rootwait [KNL] Wait (indefinitely) for root device to show up. 4892 Useful for devices that are detected asynchronously 4893 (e.g. USB and MMC devices). 4894 4895 rproc_mem=nn[KMG][@address] 4896 [KNL,ARM,CMA] Remoteproc physical memory block. 4897 Memory area to be used by remote processor image, 4898 managed by CMA. 4899 4900 rw [KNL] Mount root device read-write on boot 4901 4902 S [KNL] Run init in single mode 4903 4904 s390_iommu= [HW,S390] 4905 Set s390 IOTLB flushing mode 4906 strict 4907 With strict flushing every unmap operation will result in 4908 an IOTLB flush. Default is lazy flushing before reuse, 4909 which is faster. 4910 4911 sa1100ir [NET] 4912 See drivers/net/irda/sa1100_ir.c. 4913 4914 sbni= [NET] Granch SBNI12 leased line adapter 4915 4916 sched_debug [KNL] Enables verbose scheduler debug messages. 4917 4918 schedstats= [KNL,X86] Enable or disable scheduled statistics. 4919 Allowed values are enable and disable. This feature 4920 incurs a small amount of overhead in the scheduler 4921 but is useful for debugging and performance tuning. 4922 4923 sched_thermal_decay_shift= 4924 [KNL, SMP] Set a decay shift for scheduler thermal 4925 pressure signal. Thermal pressure signal follows the 4926 default decay period of other scheduler pelt 4927 signals(usually 32 ms but configurable). Setting 4928 sched_thermal_decay_shift will left shift the decay 4929 period for the thermal pressure signal by the shift 4930 value. 4931 i.e. with the default pelt decay period of 32 ms 4932 sched_thermal_decay_shift thermal pressure decay pr 4933 1 64 ms 4934 2 128 ms 4935 and so on. 4936 Format: integer between 0 and 10 4937 Default is 0. 4938 4939 scftorture.holdoff= [KNL] 4940 Number of seconds to hold off before starting 4941 test. Defaults to zero for module insertion and 4942 to 10 seconds for built-in smp_call_function() 4943 tests. 4944 4945 scftorture.longwait= [KNL] 4946 Request ridiculously long waits randomly selected 4947 up to the chosen limit in seconds. Zero (the 4948 default) disables this feature. Please note 4949 that requesting even small non-zero numbers of 4950 seconds can result in RCU CPU stall warnings, 4951 softlockup complaints, and so on. 4952 4953 scftorture.nthreads= [KNL] 4954 Number of kthreads to spawn to invoke the 4955 smp_call_function() family of functions. 4956 The default of -1 specifies a number of kthreads 4957 equal to the number of CPUs. 4958 4959 scftorture.onoff_holdoff= [KNL] 4960 Number seconds to wait after the start of the 4961 test before initiating CPU-hotplug operations. 4962 4963 scftorture.onoff_interval= [KNL] 4964 Number seconds to wait between successive 4965 CPU-hotplug operations. Specifying zero (which 4966 is the default) disables CPU-hotplug operations. 4967 4968 scftorture.shutdown_secs= [KNL] 4969 The number of seconds following the start of the 4970 test after which to shut down the system. The 4971 default of zero avoids shutting down the system. 4972 Non-zero values are useful for automated tests. 4973 4974 scftorture.stat_interval= [KNL] 4975 The number of seconds between outputting the 4976 current test statistics to the console. A value 4977 of zero disables statistics output. 4978 4979 scftorture.stutter_cpus= [KNL] 4980 The number of jiffies to wait between each change 4981 to the set of CPUs under test. 4982 4983 scftorture.use_cpus_read_lock= [KNL] 4984 Use use_cpus_read_lock() instead of the default 4985 preempt_disable() to disable CPU hotplug 4986 while invoking one of the smp_call_function*() 4987 functions. 4988 4989 scftorture.verbose= [KNL] 4990 Enable additional printk() statements. 4991 4992 scftorture.weight_single= [KNL] 4993 The probability weighting to use for the 4994 smp_call_function_single() function with a zero 4995 "wait" parameter. A value of -1 selects the 4996 default if all other weights are -1. However, 4997 if at least one weight has some other value, a 4998 value of -1 will instead select a weight of zero. 4999 5000 scftorture.weight_single_wait= [KNL] 5001 The probability weighting to use for the 5002 smp_call_function_single() function with a 5003 non-zero "wait" parameter. See weight_single. 5004 5005 scftorture.weight_many= [KNL] 5006 The probability weighting to use for the 5007 smp_call_function_many() function with a zero 5008 "wait" parameter. See weight_single. 5009 Note well that setting a high probability for 5010 this weighting can place serious IPI load 5011 on the system. 5012 5013 scftorture.weight_many_wait= [KNL] 5014 The probability weighting to use for the 5015 smp_call_function_many() function with a 5016 non-zero "wait" parameter. See weight_single 5017 and weight_many. 5018 5019 scftorture.weight_all= [KNL] 5020 The probability weighting to use for the 5021 smp_call_function_all() function with a zero 5022 "wait" parameter. See weight_single and 5023 weight_many. 5024 5025 scftorture.weight_all_wait= [KNL] 5026 The probability weighting to use for the 5027 smp_call_function_all() function with a 5028 non-zero "wait" parameter. See weight_single 5029 and weight_many. 5030 5031 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate 5032 xtime_lock contention on larger systems, and/or RCU lock 5033 contention on all systems with CONFIG_MAXSMP set. 5034 Format: { "0" | "1" } 5035 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1" 5036 1 -- enable. 5037 Note: increases power consumption, thus should only be 5038 enabled if running jitter sensitive (HPC/RT) workloads. 5039 5040 security= [SECURITY] Choose a legacy "major" security module to 5041 enable at boot. This has been deprecated by the 5042 "lsm=" parameter. 5043 5044 selinux= [SELINUX] Disable or enable SELinux at boot time. 5045 Format: { "0" | "1" } 5046 See security/selinux/Kconfig help text. 5047 0 -- disable. 5048 1 -- enable. 5049 Default value is 1. 5050 5051 apparmor= [APPARMOR] Disable or enable AppArmor at boot time 5052 Format: { "0" | "1" } 5053 See security/apparmor/Kconfig help text 5054 0 -- disable. 5055 1 -- enable. 5056 Default value is set via kernel config option. 5057 5058 serialnumber [BUGS=X86-32] 5059 5060 shapers= [NET] 5061 Maximal number of shapers. 5062 5063 simeth= [IA-64] 5064 simscsi= 5065 5066 slram= [HW,MTD] 5067 5068 slab_nomerge [MM] 5069 Disable merging of slabs with similar size. May be 5070 necessary if there is some reason to distinguish 5071 allocs to different slabs, especially in hardened 5072 environments where the risk of heap overflows and 5073 layout control by attackers can usually be 5074 frustrated by disabling merging. This will reduce 5075 most of the exposure of a heap attack to a single 5076 cache (risks via metadata attacks are mostly 5077 unchanged). Debug options disable merging on their 5078 own. 5079 For more information see Documentation/vm/slub.rst. 5080 5081 slab_max_order= [MM, SLAB] 5082 Determines the maximum allowed order for slabs. 5083 A high setting may cause OOMs due to memory 5084 fragmentation. Defaults to 1 for systems with 5085 more than 32MB of RAM, 0 otherwise. 5086 5087 slub_debug[=options[,slabs][;[options[,slabs]]...] [MM, SLUB] 5088 Enabling slub_debug allows one to determine the 5089 culprit if slab objects become corrupted. Enabling 5090 slub_debug can create guard zones around objects and 5091 may poison objects when not in use. Also tracks the 5092 last alloc / free. For more information see 5093 Documentation/vm/slub.rst. 5094 5095 slub_memcg_sysfs= [MM, SLUB] 5096 Determines whether to enable sysfs directories for 5097 memory cgroup sub-caches. 1 to enable, 0 to disable. 5098 The default is determined by CONFIG_SLUB_MEMCG_SYSFS_ON. 5099 Enabling this can lead to a very high number of debug 5100 directories and files being created under 5101 /sys/kernel/slub. 5102 5103 slub_max_order= [MM, SLUB] 5104 Determines the maximum allowed order for slabs. 5105 A high setting may cause OOMs due to memory 5106 fragmentation. For more information see 5107 Documentation/vm/slub.rst. 5108 5109 slub_min_objects= [MM, SLUB] 5110 The minimum number of objects per slab. SLUB will 5111 increase the slab order up to slub_max_order to 5112 generate a sufficiently large slab able to contain 5113 the number of objects indicated. The higher the number 5114 of objects the smaller the overhead of tracking slabs 5115 and the less frequently locks need to be acquired. 5116 For more information see Documentation/vm/slub.rst. 5117 5118 slub_min_order= [MM, SLUB] 5119 Determines the minimum page order for slabs. Must be 5120 lower than slub_max_order. 5121 For more information see Documentation/vm/slub.rst. 5122 5123 slub_nomerge [MM, SLUB] 5124 Same with slab_nomerge. This is supported for legacy. 5125 See slab_nomerge for more information. 5126 5127 smart2= [HW] 5128 Format: <io1>[,<io2>[,...,<io8>]] 5129 5130 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices 5131 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port 5132 smsc-ircc2.ircc_sir= [HW] SIR base I/O port 5133 smsc-ircc2.ircc_fir= [HW] FIR base I/O port 5134 smsc-ircc2.ircc_irq= [HW] IRQ line 5135 smsc-ircc2.ircc_dma= [HW] DMA channel 5136 smsc-ircc2.ircc_transceiver= [HW] Transceiver type: 5137 0: Toshiba Satellite 1800 (GP data pin select) 5138 1: Fast pin select (default) 5139 2: ATC IRMode 5140 5141 smt [KNL,S390] Set the maximum number of threads (logical 5142 CPUs) to use per physical CPU on systems capable of 5143 symmetric multithreading (SMT). Will be capped to the 5144 actual hardware limit. 5145 Format: <integer> 5146 Default: -1 (no limit) 5147 5148 softlockup_panic= 5149 [KNL] Should the soft-lockup detector generate panics. 5150 Format: 0 | 1 5151 5152 A value of 1 instructs the soft-lockup detector 5153 to panic the machine when a soft-lockup occurs. It is 5154 also controlled by the kernel.softlockup_panic sysctl 5155 and CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC, which is the 5156 respective build-time switch to that functionality. 5157 5158 softlockup_all_cpu_backtrace= 5159 [KNL] Should the soft-lockup detector generate 5160 backtraces on all cpus. 5161 Format: 0 | 1 5162 5163 sonypi.*= [HW] Sony Programmable I/O Control Device driver 5164 See Documentation/admin-guide/laptops/sonypi.rst 5165 5166 spectre_v2= [X86] Control mitigation of Spectre variant 2 5167 (indirect branch speculation) vulnerability. 5168 The default operation protects the kernel from 5169 user space attacks. 5170 5171 on - unconditionally enable, implies 5172 spectre_v2_user=on 5173 off - unconditionally disable, implies 5174 spectre_v2_user=off 5175 auto - kernel detects whether your CPU model is 5176 vulnerable 5177 5178 Selecting 'on' will, and 'auto' may, choose a 5179 mitigation method at run time according to the 5180 CPU, the available microcode, the setting of the 5181 CONFIG_RETPOLINE configuration option, and the 5182 compiler with which the kernel was built. 5183 5184 Selecting 'on' will also enable the mitigation 5185 against user space to user space task attacks. 5186 5187 Selecting 'off' will disable both the kernel and 5188 the user space protections. 5189 5190 Specific mitigations can also be selected manually: 5191 5192 retpoline - replace indirect branches 5193 retpoline,generic - Retpolines 5194 retpoline,lfence - LFENCE; indirect branch 5195 retpoline,amd - alias for retpoline,lfence 5196 eibrs - enhanced IBRS 5197 eibrs,retpoline - enhanced IBRS + Retpolines 5198 eibrs,lfence - enhanced IBRS + LFENCE 5199 ibrs - use IBRS to protect kernel 5200 5201 Not specifying this option is equivalent to 5202 spectre_v2=auto. 5203 5204 spectre_v2_user= 5205 [X86] Control mitigation of Spectre variant 2 5206 (indirect branch speculation) vulnerability between 5207 user space tasks 5208 5209 on - Unconditionally enable mitigations. Is 5210 enforced by spectre_v2=on 5211 5212 off - Unconditionally disable mitigations. Is 5213 enforced by spectre_v2=off 5214 5215 prctl - Indirect branch speculation is enabled, 5216 but mitigation can be enabled via prctl 5217 per thread. The mitigation control state 5218 is inherited on fork. 5219 5220 prctl,ibpb 5221 - Like "prctl" above, but only STIBP is 5222 controlled per thread. IBPB is issued 5223 always when switching between different user 5224 space processes. 5225 5226 seccomp 5227 - Same as "prctl" above, but all seccomp 5228 threads will enable the mitigation unless 5229 they explicitly opt out. 5230 5231 seccomp,ibpb 5232 - Like "seccomp" above, but only STIBP is 5233 controlled per thread. IBPB is issued 5234 always when switching between different 5235 user space processes. 5236 5237 auto - Kernel selects the mitigation depending on 5238 the available CPU features and vulnerability. 5239 5240 Default mitigation: 5241 If CONFIG_SECCOMP=y then "seccomp", otherwise "prctl" 5242 5243 Not specifying this option is equivalent to 5244 spectre_v2_user=auto. 5245 5246 spec_rstack_overflow= 5247 [X86] Control RAS overflow mitigation on AMD Zen CPUs 5248 5249 off - Disable mitigation 5250 microcode - Enable microcode mitigation only 5251 safe-ret - Enable sw-only safe RET mitigation (default) 5252 ibpb - Enable mitigation by issuing IBPB on 5253 kernel entry 5254 ibpb-vmexit - Issue IBPB only on VMEXIT 5255 (cloud-specific mitigation) 5256 5257 spec_store_bypass_disable= 5258 [HW] Control Speculative Store Bypass (SSB) Disable mitigation 5259 (Speculative Store Bypass vulnerability) 5260 5261 Certain CPUs are vulnerable to an exploit against a 5262 a common industry wide performance optimization known 5263 as "Speculative Store Bypass" in which recent stores 5264 to the same memory location may not be observed by 5265 later loads during speculative execution. The idea 5266 is that such stores are unlikely and that they can 5267 be detected prior to instruction retirement at the 5268 end of a particular speculation execution window. 5269 5270 In vulnerable processors, the speculatively forwarded 5271 store can be used in a cache side channel attack, for 5272 example to read memory to which the attacker does not 5273 directly have access (e.g. inside sandboxed code). 5274 5275 This parameter controls whether the Speculative Store 5276 Bypass optimization is used. 5277 5278 On x86 the options are: 5279 5280 on - Unconditionally disable Speculative Store Bypass 5281 off - Unconditionally enable Speculative Store Bypass 5282 auto - Kernel detects whether the CPU model contains an 5283 implementation of Speculative Store Bypass and 5284 picks the most appropriate mitigation. If the 5285 CPU is not vulnerable, "off" is selected. If the 5286 CPU is vulnerable the default mitigation is 5287 architecture and Kconfig dependent. See below. 5288 prctl - Control Speculative Store Bypass per thread 5289 via prctl. Speculative Store Bypass is enabled 5290 for a process by default. The state of the control 5291 is inherited on fork. 5292 seccomp - Same as "prctl" above, but all seccomp threads 5293 will disable SSB unless they explicitly opt out. 5294 5295 Default mitigations: 5296 X86: If CONFIG_SECCOMP=y "seccomp", otherwise "prctl" 5297 5298 On powerpc the options are: 5299 5300 on,auto - On Power8 and Power9 insert a store-forwarding 5301 barrier on kernel entry and exit. On Power7 5302 perform a software flush on kernel entry and 5303 exit. 5304 off - No action. 5305 5306 Not specifying this option is equivalent to 5307 spec_store_bypass_disable=auto. 5308 5309 spia_io_base= [HW,MTD] 5310 spia_fio_base= 5311 spia_pedr= 5312 spia_peddr= 5313 5314 split_lock_detect= 5315 [X86] Enable split lock detection 5316 5317 When enabled (and if hardware support is present), atomic 5318 instructions that access data across cache line 5319 boundaries will result in an alignment check exception. 5320 5321 off - not enabled 5322 5323 warn - the kernel will emit rate limited warnings 5324 about applications triggering the #AC 5325 exception. This mode is the default on CPUs 5326 that supports split lock detection. 5327 5328 fatal - the kernel will send SIGBUS to applications 5329 that trigger the #AC exception. 5330 5331 If an #AC exception is hit in the kernel or in 5332 firmware (i.e. not while executing in user mode) 5333 the kernel will oops in either "warn" or "fatal" 5334 mode. 5335 5336 srbds= [X86,INTEL] 5337 Control the Special Register Buffer Data Sampling 5338 (SRBDS) mitigation. 5339 5340 Certain CPUs are vulnerable to an MDS-like 5341 exploit which can leak bits from the random 5342 number generator. 5343 5344 By default, this issue is mitigated by 5345 microcode. However, the microcode fix can cause 5346 the RDRAND and RDSEED instructions to become 5347 much slower. Among other effects, this will 5348 result in reduced throughput from /dev/urandom. 5349 5350 The microcode mitigation can be disabled with 5351 the following option: 5352 5353 off: Disable mitigation and remove 5354 performance impact to RDRAND and RDSEED 5355 5356 srcutree.counter_wrap_check [KNL] 5357 Specifies how frequently to check for 5358 grace-period sequence counter wrap for the 5359 srcu_data structure's ->srcu_gp_seq_needed field. 5360 The greater the number of bits set in this kernel 5361 parameter, the less frequently counter wrap will 5362 be checked for. Note that the bottom two bits 5363 are ignored. 5364 5365 srcutree.exp_holdoff [KNL] 5366 Specifies how many nanoseconds must elapse 5367 since the end of the last SRCU grace period for 5368 a given srcu_struct until the next normal SRCU 5369 grace period will be considered for automatic 5370 expediting. Set to zero to disable automatic 5371 expediting. 5372 5373 ssbd= [ARM64,HW] 5374 Speculative Store Bypass Disable control 5375 5376 On CPUs that are vulnerable to the Speculative 5377 Store Bypass vulnerability and offer a 5378 firmware based mitigation, this parameter 5379 indicates how the mitigation should be used: 5380 5381 force-on: Unconditionally enable mitigation for 5382 for both kernel and userspace 5383 force-off: Unconditionally disable mitigation for 5384 for both kernel and userspace 5385 kernel: Always enable mitigation in the 5386 kernel, and offer a prctl interface 5387 to allow userspace to register its 5388 interest in being mitigated too. 5389 5390 stack_guard_gap= [MM] 5391 override the default stack gap protection. The value 5392 is in page units and it defines how many pages prior 5393 to (for stacks growing down) resp. after (for stacks 5394 growing up) the main stack are reserved for no other 5395 mapping. Default value is 256 pages. 5396 5397 stack_depot_disable= [KNL] 5398 Setting this to true through kernel command line will 5399 disable the stack depot thereby saving the static memory 5400 consumed by the stack hash table. By default this is set 5401 to false. 5402 5403 stacktrace [FTRACE] 5404 Enabled the stack tracer on boot up. 5405 5406 stacktrace_filter=[function-list] 5407 [FTRACE] Limit the functions that the stack tracer 5408 will trace at boot up. function-list is a comma separated 5409 list of functions. This list can be changed at run 5410 time by the stack_trace_filter file in the debugfs 5411 tracing directory. Note, this enables stack tracing 5412 and the stacktrace above is not needed. 5413 5414 sti= [PARISC,HW] 5415 Format: <num> 5416 Set the STI (builtin display/keyboard on the HP-PARISC 5417 machines) console (graphic card) which should be used 5418 as the initial boot-console. 5419 See also comment in drivers/video/console/sticore.c. 5420 5421 sti_font= [HW] 5422 See comment in drivers/video/console/sticore.c. 5423 5424 stifb= [HW] 5425 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]] 5426 5427 sunrpc.min_resvport= 5428 sunrpc.max_resvport= 5429 [NFS,SUNRPC] 5430 SunRPC servers often require that client requests 5431 originate from a privileged port (i.e. a port in the 5432 range 0 < portnr < 1024). 5433 An administrator who wishes to reserve some of these 5434 ports for other uses may adjust the range that the 5435 kernel's sunrpc client considers to be privileged 5436 using these two parameters to set the minimum and 5437 maximum port values. 5438 5439 sunrpc.svc_rpc_per_connection_limit= 5440 [NFS,SUNRPC] 5441 Limit the number of requests that the server will 5442 process in parallel from a single connection. 5443 The default value is 0 (no limit). 5444 5445 sunrpc.pool_mode= 5446 [NFS] 5447 Control how the NFS server code allocates CPUs to 5448 service thread pools. Depending on how many NICs 5449 you have and where their interrupts are bound, this 5450 option will affect which CPUs will do NFS serving. 5451 Note: this parameter cannot be changed while the 5452 NFS server is running. 5453 5454 auto the server chooses an appropriate mode 5455 automatically using heuristics 5456 global a single global pool contains all CPUs 5457 percpu one pool for each CPU 5458 pernode one pool for each NUMA node (equivalent 5459 to global on non-NUMA machines) 5460 5461 sunrpc.tcp_slot_table_entries= 5462 sunrpc.udp_slot_table_entries= 5463 [NFS,SUNRPC] 5464 Sets the upper limit on the number of simultaneous 5465 RPC calls that can be sent from the client to a 5466 server. Increasing these values may allow you to 5467 improve throughput, but will also increase the 5468 amount of memory reserved for use by the client. 5469 5470 suspend.pm_test_delay= 5471 [SUSPEND] 5472 Sets the number of seconds to remain in a suspend test 5473 mode before resuming the system (see 5474 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG 5475 is set. Default value is 5. 5476 5477 svm= [PPC] 5478 Format: { on | off | y | n | 1 | 0 } 5479 This parameter controls use of the Protected 5480 Execution Facility on pSeries. 5481 5482 swapaccount=[0|1] 5483 [KNL] Enable accounting of swap in memory resource 5484 controller if no parameter or 1 is given or disable 5485 it if 0 is given (See Documentation/admin-guide/cgroup-v1/memory.rst) 5486 5487 swiotlb= [ARM,IA-64,PPC,MIPS,X86] 5488 Format: { <int> | force | noforce } 5489 <int> -- Number of I/O TLB slabs 5490 force -- force using of bounce buffers even if they 5491 wouldn't be automatically used by the kernel 5492 noforce -- Never use bounce buffers (for debugging) 5493 5494 switches= [HW,M68k] 5495 5496 sysctl.*= [KNL] 5497 Set a sysctl parameter, right before loading the init 5498 process, as if the value was written to the respective 5499 /proc/sys/... file. Both '.' and '/' are recognized as 5500 separators. Unrecognized parameters and invalid values 5501 are reported in the kernel log. Sysctls registered 5502 later by a loaded module cannot be set this way. 5503 Example: sysctl.vm.swappiness=40 5504 5505 sysfs.deprecated=0|1 [KNL] 5506 Enable/disable old style sysfs layout for old udev 5507 on older distributions. When this option is enabled 5508 very new udev will not work anymore. When this option 5509 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled) 5510 in older udev will not work anymore. 5511 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in 5512 the kernel configuration. 5513 5514 sysrq_always_enabled 5515 [KNL] 5516 Ignore sysrq setting - this boot parameter will 5517 neutralize any effect of /proc/sys/kernel/sysrq. 5518 Useful for debugging. 5519 5520 tcpmhash_entries= [KNL,NET] 5521 Set the number of tcp_metrics_hash slots. 5522 Default value is 8192 or 16384 depending on total 5523 ram pages. This is used to specify the TCP metrics 5524 cache size. See Documentation/networking/ip-sysctl.rst 5525 "tcp_no_metrics_save" section for more details. 5526 5527 tdfx= [HW,DRM] 5528 5529 test_suspend= [SUSPEND][,N] 5530 Specify "mem" (for Suspend-to-RAM) or "standby" (for 5531 standby suspend) or "freeze" (for suspend type freeze) 5532 as the system sleep state during system startup with 5533 the optional capability to repeat N number of times. 5534 The system is woken from this state using a 5535 wakeup-capable RTC alarm. 5536 5537 thash_entries= [KNL,NET] 5538 Set number of hash buckets for TCP connection 5539 5540 thermal.act= [HW,ACPI] 5541 -1: disable all active trip points in all thermal zones 5542 <degrees C>: override all lowest active trip points 5543 5544 thermal.crt= [HW,ACPI] 5545 -1: disable all critical trip points in all thermal zones 5546 <degrees C>: override all critical trip points 5547 5548 thermal.nocrt= [HW,ACPI] 5549 Set to disable actions on ACPI thermal zone 5550 critical and hot trip points. 5551 5552 thermal.off= [HW,ACPI] 5553 1: disable ACPI thermal control 5554 5555 thermal.psv= [HW,ACPI] 5556 -1: disable all passive trip points 5557 <degrees C>: override all passive trip points to this 5558 value 5559 5560 thermal.tzp= [HW,ACPI] 5561 Specify global default ACPI thermal zone polling rate 5562 <deci-seconds>: poll all this frequency 5563 0: no polling (default) 5564 5565 threadirqs [KNL] 5566 Force threading of all interrupt handlers except those 5567 marked explicitly IRQF_NO_THREAD. 5568 5569 topology= [S390] 5570 Format: {off | on} 5571 Specify if the kernel should make use of the cpu 5572 topology information if the hardware supports this. 5573 The scheduler will make use of this information and 5574 e.g. base its process migration decisions on it. 5575 Default is on. 5576 5577 topology_updates= [KNL, PPC, NUMA] 5578 Format: {off} 5579 Specify if the kernel should ignore (off) 5580 topology updates sent by the hypervisor to this 5581 LPAR. 5582 5583 torture.disable_onoff_at_boot= [KNL] 5584 Prevent the CPU-hotplug component of torturing 5585 until after init has spawned. 5586 5587 torture.ftrace_dump_at_shutdown= [KNL] 5588 Dump the ftrace buffer at torture-test shutdown, 5589 even if there were no errors. This can be a 5590 very costly operation when many torture tests 5591 are running concurrently, especially on systems 5592 with rotating-rust storage. 5593 5594 tp720= [HW,PS2] 5595 5596 tpm_suspend_pcr=[HW,TPM] 5597 Format: integer pcr id 5598 Specify that at suspend time, the tpm driver 5599 should extend the specified pcr with zeros, 5600 as a workaround for some chips which fail to 5601 flush the last written pcr on TPM_SaveState. 5602 This will guarantee that all the other pcrs 5603 are saved. 5604 5605 trace_buf_size=nn[KMG] 5606 [FTRACE] will set tracing buffer size on each cpu. 5607 5608 trace_event=[event-list] 5609 [FTRACE] Set and start specified trace events in order 5610 to facilitate early boot debugging. The event-list is a 5611 comma separated list of trace events to enable. See 5612 also Documentation/trace/events.rst 5613 5614 trace_options=[option-list] 5615 [FTRACE] Enable or disable tracer options at boot. 5616 The option-list is a comma delimited list of options 5617 that can be enabled or disabled just as if you were 5618 to echo the option name into 5619 5620 /sys/kernel/debug/tracing/trace_options 5621 5622 For example, to enable stacktrace option (to dump the 5623 stack trace of each event), add to the command line: 5624 5625 trace_options=stacktrace 5626 5627 See also Documentation/trace/ftrace.rst "trace options" 5628 section. 5629 5630 tp_printk[FTRACE] 5631 Have the tracepoints sent to printk as well as the 5632 tracing ring buffer. This is useful for early boot up 5633 where the system hangs or reboots and does not give the 5634 option for reading the tracing buffer or performing a 5635 ftrace_dump_on_oops. 5636 5637 To turn off having tracepoints sent to printk, 5638 echo 0 > /proc/sys/kernel/tracepoint_printk 5639 Note, echoing 1 into this file without the 5640 tracepoint_printk kernel cmdline option has no effect. 5641 5642 ** CAUTION ** 5643 5644 Having tracepoints sent to printk() and activating high 5645 frequency tracepoints such as irq or sched, can cause 5646 the system to live lock. 5647 5648 traceoff_on_warning 5649 [FTRACE] enable this option to disable tracing when a 5650 warning is hit. This turns off "tracing_on". Tracing can 5651 be enabled again by echoing '1' into the "tracing_on" 5652 file located in /sys/kernel/debug/tracing/ 5653 5654 This option is useful, as it disables the trace before 5655 the WARNING dump is called, which prevents the trace to 5656 be filled with content caused by the warning output. 5657 5658 This option can also be set at run time via the sysctl 5659 option: kernel/traceoff_on_warning 5660 5661 transparent_hugepage= 5662 [KNL] 5663 Format: [always|madvise|never] 5664 Can be used to control the default behavior of the system 5665 with respect to transparent hugepages. 5666 See Documentation/admin-guide/mm/transhuge.rst 5667 for more details. 5668 5669 tsc= Disable clocksource stability checks for TSC. 5670 Format: <string> 5671 [x86] reliable: mark tsc clocksource as reliable, this 5672 disables clocksource verification at runtime, as well 5673 as the stability checks done at bootup. Used to enable 5674 high-resolution timer mode on older hardware, and in 5675 virtualized environment. 5676 [x86] noirqtime: Do not use TSC to do irq accounting. 5677 Used to run time disable IRQ_TIME_ACCOUNTING on any 5678 platforms where RDTSC is slow and this accounting 5679 can add overhead. 5680 [x86] unstable: mark the TSC clocksource as unstable, this 5681 marks the TSC unconditionally unstable at bootup and 5682 avoids any further wobbles once the TSC watchdog notices. 5683 [x86] nowatchdog: disable clocksource watchdog. Used 5684 in situations with strict latency requirements (where 5685 interruptions from clocksource watchdog are not 5686 acceptable). 5687 5688 tsc_early_khz= [X86] Skip early TSC calibration and use the given 5689 value instead. Useful when the early TSC frequency discovery 5690 procedure is not reliable, such as on overclocked systems 5691 with CPUID.16h support and partial CPUID.15h support. 5692 Format: <unsigned int> 5693 5694 tsx= [X86] Control Transactional Synchronization 5695 Extensions (TSX) feature in Intel processors that 5696 support TSX control. 5697 5698 This parameter controls the TSX feature. The options are: 5699 5700 on - Enable TSX on the system. Although there are 5701 mitigations for all known security vulnerabilities, 5702 TSX has been known to be an accelerator for 5703 several previous speculation-related CVEs, and 5704 so there may be unknown security risks associated 5705 with leaving it enabled. 5706 5707 off - Disable TSX on the system. (Note that this 5708 option takes effect only on newer CPUs which are 5709 not vulnerable to MDS, i.e., have 5710 MSR_IA32_ARCH_CAPABILITIES.MDS_NO=1 and which get 5711 the new IA32_TSX_CTRL MSR through a microcode 5712 update. This new MSR allows for the reliable 5713 deactivation of the TSX functionality.) 5714 5715 auto - Disable TSX if X86_BUG_TAA is present, 5716 otherwise enable TSX on the system. 5717 5718 Not specifying this option is equivalent to tsx=off. 5719 5720 See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst 5721 for more details. 5722 5723 tsx_async_abort= [X86,INTEL] Control mitigation for the TSX Async 5724 Abort (TAA) vulnerability. 5725 5726 Similar to Micro-architectural Data Sampling (MDS) 5727 certain CPUs that support Transactional 5728 Synchronization Extensions (TSX) are vulnerable to an 5729 exploit against CPU internal buffers which can forward 5730 information to a disclosure gadget under certain 5731 conditions. 5732 5733 In vulnerable processors, the speculatively forwarded 5734 data can be used in a cache side channel attack, to 5735 access data to which the attacker does not have direct 5736 access. 5737 5738 This parameter controls the TAA mitigation. The 5739 options are: 5740 5741 full - Enable TAA mitigation on vulnerable CPUs 5742 if TSX is enabled. 5743 5744 full,nosmt - Enable TAA mitigation and disable SMT on 5745 vulnerable CPUs. If TSX is disabled, SMT 5746 is not disabled because CPU is not 5747 vulnerable to cross-thread TAA attacks. 5748 off - Unconditionally disable TAA mitigation 5749 5750 On MDS-affected machines, tsx_async_abort=off can be 5751 prevented by an active MDS mitigation as both vulnerabilities 5752 are mitigated with the same mechanism so in order to disable 5753 this mitigation, you need to specify mds=off too. 5754 5755 Not specifying this option is equivalent to 5756 tsx_async_abort=full. On CPUs which are MDS affected 5757 and deploy MDS mitigation, TAA mitigation is not 5758 required and doesn't provide any additional 5759 mitigation. 5760 5761 For details see: 5762 Documentation/admin-guide/hw-vuln/tsx_async_abort.rst 5763 5764 turbografx.map[2|3]= [HW,JOY] 5765 TurboGraFX parallel port interface 5766 Format: 5767 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7> 5768 See also Documentation/input/devices/joystick-parport.rst 5769 5770 udbg-immortal [PPC] When debugging early kernel crashes that 5771 happen after console_init() and before a proper 5772 console driver takes over, this boot options might 5773 help "seeing" what's going on. 5774 5775 uhash_entries= [KNL,NET] 5776 Set number of hash buckets for UDP/UDP-Lite connections 5777 5778 uhci-hcd.ignore_oc= 5779 [USB] Ignore overcurrent events (default N). 5780 Some badly-designed motherboards generate lots of 5781 bogus events, for ports that aren't wired to 5782 anything. Set this parameter to avoid log spamming. 5783 Note that genuine overcurrent events won't be 5784 reported either. 5785 5786 unknown_nmi_panic 5787 [X86] Cause panic on unknown NMI. 5788 5789 usbcore.authorized_default= 5790 [USB] Default USB device authorization: 5791 (default -1 = authorized except for wireless USB, 5792 0 = not authorized, 1 = authorized, 2 = authorized 5793 if device connected to internal port) 5794 5795 usbcore.autosuspend= 5796 [USB] The autosuspend time delay (in seconds) used 5797 for newly-detected USB devices (default 2). This 5798 is the time required before an idle device will be 5799 autosuspended. Devices for which the delay is set 5800 to a negative value won't be autosuspended at all. 5801 5802 usbcore.usbfs_snoop= 5803 [USB] Set to log all usbfs traffic (default 0 = off). 5804 5805 usbcore.usbfs_snoop_max= 5806 [USB] Maximum number of bytes to snoop in each URB 5807 (default = 65536). 5808 5809 usbcore.blinkenlights= 5810 [USB] Set to cycle leds on hubs (default 0 = off). 5811 5812 usbcore.old_scheme_first= 5813 [USB] Start with the old device initialization 5814 scheme (default 0 = off). 5815 5816 usbcore.usbfs_memory_mb= 5817 [USB] Memory limit (in MB) for buffers allocated by 5818 usbfs (default = 16, 0 = max = 2047). 5819 5820 usbcore.use_both_schemes= 5821 [USB] Try the other device initialization scheme 5822 if the first one fails (default 1 = enabled). 5823 5824 usbcore.initial_descriptor_timeout= 5825 [USB] Specifies timeout for the initial 64-byte 5826 USB_REQ_GET_DESCRIPTOR request in milliseconds 5827 (default 5000 = 5.0 seconds). 5828 5829 usbcore.nousb [USB] Disable the USB subsystem 5830 5831 usbcore.quirks= 5832 [USB] A list of quirk entries to augment the built-in 5833 usb core quirk list. List entries are separated by 5834 commas. Each entry has the form 5835 VendorID:ProductID:Flags. The IDs are 4-digit hex 5836 numbers and Flags is a set of letters. Each letter 5837 will change the built-in quirk; setting it if it is 5838 clear and clearing it if it is set. The letters have 5839 the following meanings: 5840 a = USB_QUIRK_STRING_FETCH_255 (string 5841 descriptors must not be fetched using 5842 a 255-byte read); 5843 b = USB_QUIRK_RESET_RESUME (device can't resume 5844 correctly so reset it instead); 5845 c = USB_QUIRK_NO_SET_INTF (device can't handle 5846 Set-Interface requests); 5847 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't 5848 handle its Configuration or Interface 5849 strings); 5850 e = USB_QUIRK_RESET (device can't be reset 5851 (e.g morph devices), don't use reset); 5852 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has 5853 more interface descriptions than the 5854 bNumInterfaces count, and can't handle 5855 talking to these interfaces); 5856 g = USB_QUIRK_DELAY_INIT (device needs a pause 5857 during initialization, after we read 5858 the device descriptor); 5859 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For 5860 high speed and super speed interrupt 5861 endpoints, the USB 2.0 and USB 3.0 spec 5862 require the interval in microframes (1 5863 microframe = 125 microseconds) to be 5864 calculated as interval = 2 ^ 5865 (bInterval-1). 5866 Devices with this quirk report their 5867 bInterval as the result of this 5868 calculation instead of the exponent 5869 variable used in the calculation); 5870 i = USB_QUIRK_DEVICE_QUALIFIER (device can't 5871 handle device_qualifier descriptor 5872 requests); 5873 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device 5874 generates spurious wakeup, ignore 5875 remote wakeup capability); 5876 k = USB_QUIRK_NO_LPM (device can't handle Link 5877 Power Management); 5878 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL 5879 (Device reports its bInterval as linear 5880 frames instead of the USB 2.0 5881 calculation); 5882 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs 5883 to be disconnected before suspend to 5884 prevent spurious wakeup); 5885 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a 5886 pause after every control message); 5887 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra 5888 delay after resetting its port); 5889 Example: quirks=0781:5580:bk,0a5c:5834:gij 5890 5891 usbhid.mousepoll= 5892 [USBHID] The interval which mice are to be polled at. 5893 5894 usbhid.jspoll= 5895 [USBHID] The interval which joysticks are to be polled at. 5896 5897 usbhid.kbpoll= 5898 [USBHID] The interval which keyboards are to be polled at. 5899 5900 usb-storage.delay_use= 5901 [UMS] The delay in seconds before a new device is 5902 scanned for Logical Units (default 1). 5903 5904 usb-storage.quirks= 5905 [UMS] A list of quirks entries to supplement or 5906 override the built-in unusual_devs list. List 5907 entries are separated by commas. Each entry has 5908 the form VID:PID:Flags where VID and PID are Vendor 5909 and Product ID values (4-digit hex numbers) and 5910 Flags is a set of characters, each corresponding 5911 to a common usb-storage quirk flag as follows: 5912 a = SANE_SENSE (collect more than 18 bytes 5913 of sense data, not on uas); 5914 b = BAD_SENSE (don't collect more than 18 5915 bytes of sense data, not on uas); 5916 c = FIX_CAPACITY (decrease the reported 5917 device capacity by one sector); 5918 d = NO_READ_DISC_INFO (don't use 5919 READ_DISC_INFO command, not on uas); 5920 e = NO_READ_CAPACITY_16 (don't use 5921 READ_CAPACITY_16 command); 5922 f = NO_REPORT_OPCODES (don't use report opcodes 5923 command, uas only); 5924 g = MAX_SECTORS_240 (don't transfer more than 5925 240 sectors at a time, uas only); 5926 h = CAPACITY_HEURISTICS (decrease the 5927 reported device capacity by one 5928 sector if the number is odd); 5929 i = IGNORE_DEVICE (don't bind to this 5930 device); 5931 j = NO_REPORT_LUNS (don't use report luns 5932 command, uas only); 5933 k = NO_SAME (do not use WRITE_SAME, uas only) 5934 l = NOT_LOCKABLE (don't try to lock and 5935 unlock ejectable media, not on uas); 5936 m = MAX_SECTORS_64 (don't transfer more 5937 than 64 sectors = 32 KB at a time, 5938 not on uas); 5939 n = INITIAL_READ10 (force a retry of the 5940 initial READ(10) command, not on uas); 5941 o = CAPACITY_OK (accept the capacity 5942 reported by the device, not on uas); 5943 p = WRITE_CACHE (the device cache is ON 5944 by default, not on uas); 5945 r = IGNORE_RESIDUE (the device reports 5946 bogus residue values, not on uas); 5947 s = SINGLE_LUN (the device has only one 5948 Logical Unit); 5949 t = NO_ATA_1X (don't allow ATA(12) and ATA(16) 5950 commands, uas only); 5951 u = IGNORE_UAS (don't bind to the uas driver); 5952 w = NO_WP_DETECT (don't test whether the 5953 medium is write-protected). 5954 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE 5955 even if the device claims no cache, 5956 not on uas) 5957 Example: quirks=0419:aaf5:rl,0421:0433:rc 5958 5959 user_debug= [KNL,ARM] 5960 Format: <int> 5961 See arch/arm/Kconfig.debug help text. 5962 1 - undefined instruction events 5963 2 - system calls 5964 4 - invalid data aborts 5965 8 - SIGSEGV faults 5966 16 - SIGBUS faults 5967 Example: user_debug=31 5968 5969 userpte= 5970 [X86] Flags controlling user PTE allocations. 5971 5972 nohigh = do not allocate PTE pages in 5973 HIGHMEM regardless of setting 5974 of CONFIG_HIGHPTE. 5975 5976 vdso= [X86,SH] 5977 On X86_32, this is an alias for vdso32=. Otherwise: 5978 5979 vdso=1: enable VDSO (the default) 5980 vdso=0: disable VDSO mapping 5981 5982 vdso32= [X86] Control the 32-bit vDSO 5983 vdso32=1: enable 32-bit VDSO 5984 vdso32=0 or vdso32=2: disable 32-bit VDSO 5985 5986 See the help text for CONFIG_COMPAT_VDSO for more 5987 details. If CONFIG_COMPAT_VDSO is set, the default is 5988 vdso32=0; otherwise, the default is vdso32=1. 5989 5990 For compatibility with older kernels, vdso32=2 is an 5991 alias for vdso32=0. 5992 5993 Try vdso32=0 if you encounter an error that says: 5994 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed! 5995 5996 vector= [IA-64,SMP] 5997 vector=percpu: enable percpu vector domain 5998 5999 video= [FB] Frame buffer configuration 6000 See Documentation/fb/modedb.rst. 6001 6002 video.brightness_switch_enabled= [0,1] 6003 If set to 1, on receiving an ACPI notify event 6004 generated by hotkey, video driver will adjust brightness 6005 level and then send out the event to user space through 6006 the allocated input device; If set to 0, video driver 6007 will only send out the event without touching backlight 6008 brightness level. 6009 default: 1 6010 6011 virtio_mmio.device= 6012 [VMMIO] Memory mapped virtio (platform) device. 6013 6014 <size>@<baseaddr>:<irq>[:<id>] 6015 where: 6016 <size> := size (can use standard suffixes 6017 like K, M and G) 6018 <baseaddr> := physical base address 6019 <irq> := interrupt number (as passed to 6020 request_irq()) 6021 <id> := (optional) platform device id 6022 example: 6023 virtio_mmio.device=1K@0x100b0000:48:7 6024 6025 Can be used multiple times for multiple devices. 6026 6027 vga= [BOOT,X86-32] Select a particular video mode 6028 See Documentation/x86/boot.rst and 6029 Documentation/admin-guide/svga.rst. 6030 Use vga=ask for menu. 6031 This is actually a boot loader parameter; the value is 6032 passed to the kernel using a special protocol. 6033 6034 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y. 6035 May slow down system boot speed, especially when 6036 enabled on systems with a large amount of memory. 6037 All options are enabled by default, and this 6038 interface is meant to allow for selectively 6039 enabling or disabling specific virtual memory 6040 debugging features. 6041 6042 Available options are: 6043 P Enable page structure init time poisoning 6044 - Disable all of the above options 6045 6046 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact 6047 size of <nn>. This can be used to increase the 6048 minimum size (128MB on x86). It can also be used to 6049 decrease the size and leave more room for directly 6050 mapped kernel RAM. 6051 6052 vmcp_cma=nn[MG] [KNL,S390] 6053 Sets the memory size reserved for contiguous memory 6054 allocations for the vmcp device driver. 6055 6056 vmhalt= [KNL,S390] Perform z/VM CP command after system halt. 6057 Format: <command> 6058 6059 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic. 6060 Format: <command> 6061 6062 vmpoff= [KNL,S390] Perform z/VM CP command after power off. 6063 Format: <command> 6064 6065 vsyscall= [X86-64] 6066 Controls the behavior of vsyscalls (i.e. calls to 6067 fixed addresses of 0xffffffffff600x00 from legacy 6068 code). Most statically-linked binaries and older 6069 versions of glibc use these calls. Because these 6070 functions are at fixed addresses, they make nice 6071 targets for exploits that can control RIP. 6072 6073 emulate [default] Vsyscalls turn into traps and are 6074 emulated reasonably safely. The vsyscall 6075 page is readable. 6076 6077 xonly Vsyscalls turn into traps and are 6078 emulated reasonably safely. The vsyscall 6079 page is not readable. 6080 6081 none Vsyscalls don't work at all. This makes 6082 them quite hard to use for exploits but 6083 might break your system. 6084 6085 vt.color= [VT] Default text color. 6086 Format: 0xYX, X = foreground, Y = background. 6087 Default: 0x07 = light gray on black. 6088 6089 vt.cur_default= [VT] Default cursor shape. 6090 Format: 0xCCBBAA, where AA, BB, and CC are the same as 6091 the parameters of the <Esc>[?A;B;Cc escape sequence; 6092 see VGA-softcursor.txt. Default: 2 = underline. 6093 6094 vt.default_blu= [VT] 6095 Format: <blue0>,<blue1>,<blue2>,...,<blue15> 6096 Change the default blue palette of the console. 6097 This is a 16-member array composed of values 6098 ranging from 0-255. 6099 6100 vt.default_grn= [VT] 6101 Format: <green0>,<green1>,<green2>,...,<green15> 6102 Change the default green palette of the console. 6103 This is a 16-member array composed of values 6104 ranging from 0-255. 6105 6106 vt.default_red= [VT] 6107 Format: <red0>,<red1>,<red2>,...,<red15> 6108 Change the default red palette of the console. 6109 This is a 16-member array composed of values 6110 ranging from 0-255. 6111 6112 vt.default_utf8= 6113 [VT] 6114 Format=<0|1> 6115 Set system-wide default UTF-8 mode for all tty's. 6116 Default is 1, i.e. UTF-8 mode is enabled for all 6117 newly opened terminals. 6118 6119 vt.global_cursor_default= 6120 [VT] 6121 Format=<-1|0|1> 6122 Set system-wide default for whether a cursor 6123 is shown on new VTs. Default is -1, 6124 i.e. cursors will be created by default unless 6125 overridden by individual drivers. 0 will hide 6126 cursors, 1 will display them. 6127 6128 vt.italic= [VT] Default color for italic text; 0-15. 6129 Default: 2 = green. 6130 6131 vt.underline= [VT] Default color for underlined text; 0-15. 6132 Default: 3 = cyan. 6133 6134 watchdog timers [HW,WDT] For information on watchdog timers, 6135 see Documentation/watchdog/watchdog-parameters.rst 6136 or other driver-specific files in the 6137 Documentation/watchdog/ directory. 6138 6139 watchdog_thresh= 6140 [KNL] 6141 Set the hard lockup detector stall duration 6142 threshold in seconds. The soft lockup detector 6143 threshold is set to twice the value. A value of 0 6144 disables both lockup detectors. Default is 10 6145 seconds. 6146 6147 workqueue.watchdog_thresh= 6148 If CONFIG_WQ_WATCHDOG is configured, workqueue can 6149 warn stall conditions and dump internal state to 6150 help debugging. 0 disables workqueue stall 6151 detection; otherwise, it's the stall threshold 6152 duration in seconds. The default value is 30 and 6153 it can be updated at runtime by writing to the 6154 corresponding sysfs file. 6155 6156 workqueue.disable_numa 6157 By default, all work items queued to unbound 6158 workqueues are affine to the NUMA nodes they're 6159 issued on, which results in better behavior in 6160 general. If NUMA affinity needs to be disabled for 6161 whatever reason, this option can be used. Note 6162 that this also can be controlled per-workqueue for 6163 workqueues visible under /sys/bus/workqueue/. 6164 6165 workqueue.power_efficient 6166 Per-cpu workqueues are generally preferred because 6167 they show better performance thanks to cache 6168 locality; unfortunately, per-cpu workqueues tend to 6169 be more power hungry than unbound workqueues. 6170 6171 Enabling this makes the per-cpu workqueues which 6172 were observed to contribute significantly to power 6173 consumption unbound, leading to measurably lower 6174 power usage at the cost of small performance 6175 overhead. 6176 6177 The default value of this parameter is determined by 6178 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT. 6179 6180 workqueue.debug_force_rr_cpu 6181 Workqueue used to implicitly guarantee that work 6182 items queued without explicit CPU specified are put 6183 on the local CPU. This guarantee is no longer true 6184 and while local CPU is still preferred work items 6185 may be put on foreign CPUs. This debug option 6186 forces round-robin CPU selection to flush out 6187 usages which depend on the now broken guarantee. 6188 When enabled, memory and cache locality will be 6189 impacted. 6190 6191 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of 6192 default x2apic cluster mode on platforms 6193 supporting x2apic. 6194 6195 x86_intel_mid_timer= [X86-32,APBT] 6196 Choose timer option for x86 Intel MID platform. 6197 Two valid options are apbt timer only and lapic timer 6198 plus one apbt timer for broadcast timer. 6199 x86_intel_mid_timer=apbt_only | lapic_and_apbt 6200 6201 xen_512gb_limit [KNL,X86-64,XEN] 6202 Restricts the kernel running paravirtualized under Xen 6203 to use only up to 512 GB of RAM. The reason to do so is 6204 crash analysis tools and Xen tools for doing domain 6205 save/restore/migration must be enabled to handle larger 6206 domains. 6207 6208 xen_emul_unplug= [HW,X86,XEN] 6209 Unplug Xen emulated devices 6210 Format: [unplug0,][unplug1] 6211 ide-disks -- unplug primary master IDE devices 6212 aux-ide-disks -- unplug non-primary-master IDE devices 6213 nics -- unplug network devices 6214 all -- unplug all emulated devices (NICs and IDE disks) 6215 unnecessary -- unplugging emulated devices is 6216 unnecessary even if the host did not respond to 6217 the unplug protocol 6218 never -- do not unplug even if version check succeeds 6219 6220 xen_legacy_crash [X86,XEN] 6221 Crash from Xen panic notifier, without executing late 6222 panic() code such as dumping handler. 6223 6224 xen_nopvspin [X86,XEN] 6225 Disables the qspinlock slowpath using Xen PV optimizations. 6226 This parameter is obsoleted by "nopvspin" parameter, which 6227 has equivalent effect for XEN platform. 6228 6229 xen_nopv [X86] 6230 Disables the PV optimizations forcing the HVM guest to 6231 run as generic HVM guest with no PV drivers. 6232 This option is obsoleted by the "nopv" option, which 6233 has equivalent effect for XEN platform. 6234 6235 xen_no_vector_callback 6236 [KNL,X86,XEN] Disable the vector callback for Xen 6237 event channel interrupts. 6238 6239 xen_scrub_pages= [XEN] 6240 Boolean option to control scrubbing pages before giving them back 6241 to Xen, for use by other domains. Can be also changed at runtime 6242 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages. 6243 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT. 6244 6245 xen_timer_slop= [X86-64,XEN] 6246 Set the timer slop (in nanoseconds) for the virtual Xen 6247 timers (default is 100000). This adjusts the minimum 6248 delta of virtualized Xen timers, where lower values 6249 improve timer resolution at the expense of processing 6250 more timer interrupts. 6251 6252 xen.balloon_boot_timeout= [XEN] 6253 The time (in seconds) to wait before giving up to boot 6254 in case initial ballooning fails to free enough memory. 6255 Applies only when running as HVM or PVH guest and 6256 started with less memory configured than allowed at 6257 max. Default is 180. 6258 6259 xen.event_eoi_delay= [XEN] 6260 How long to delay EOI handling in case of event 6261 storms (jiffies). Default is 10. 6262 6263 xen.event_loop_timeout= [XEN] 6264 After which time (jiffies) the event handling loop 6265 should start to delay EOI handling. Default is 2. 6266 6267 xen.fifo_events= [XEN] 6268 Boolean parameter to disable using fifo event handling 6269 even if available. Normally fifo event handling is 6270 preferred over the 2-level event handling, as it is 6271 fairer and the number of possible event channels is 6272 much higher. Default is on (use fifo events). 6273 6274 nopv= [X86,XEN,KVM,HYPER_V,VMWARE] 6275 Disables the PV optimizations forcing the guest to run 6276 as generic guest with no PV drivers. Currently support 6277 XEN HVM, KVM, HYPER_V and VMWARE guest. 6278 6279 nopvspin [X86,XEN,KVM] 6280 Disables the qspinlock slow path using PV optimizations 6281 which allow the hypervisor to 'idle' the guest on lock 6282 contention. 6283 6284 xirc2ps_cs= [NET,PCMCIA] 6285 Format: 6286 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]] 6287 6288 xive= [PPC] 6289 By default on POWER9 and above, the kernel will 6290 natively use the XIVE interrupt controller. This option 6291 allows the fallback firmware mode to be used: 6292 6293 off Fallback to firmware control of XIVE interrupt 6294 controller on both pseries and powernv 6295 platforms. Only useful on POWER9 and above. 6296 6297 xhci-hcd.quirks [USB,KNL] 6298 A hex value specifying bitmask with supplemental xhci 6299 host controller quirks. Meaning of each bit can be 6300 consulted in header drivers/usb/host/xhci.h. 6301 6302 xmon [PPC] 6303 Format: { early | on | rw | ro | off } 6304 Controls if xmon debugger is enabled. Default is off. 6305 Passing only "xmon" is equivalent to "xmon=early". 6306 early Call xmon as early as possible on boot; xmon 6307 debugger is called from setup_arch(). 6308 on xmon debugger hooks will be installed so xmon 6309 is only called on a kernel crash. Default mode, 6310 i.e. either "ro" or "rw" mode, is controlled 6311 with CONFIG_XMON_DEFAULT_RO_MODE. 6312 rw xmon debugger hooks will be installed so xmon 6313 is called only on a kernel crash, mode is write, 6314 meaning SPR registers, memory and, other data 6315 can be written using xmon commands. 6316 ro same as "rw" option above but SPR registers, 6317 memory, and other data can't be written using 6318 xmon commands. 6319 off xmon is disabled. 6320