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 amd_iommu= [HW,X86-64] 293 Pass parameters to the AMD IOMMU driver in the system. 294 Possible values are: 295 fullflush - enable flushing of IO/TLB entries when 296 they are unmapped. Otherwise they are 297 flushed before they will be reused, which 298 is a lot of faster 299 off - do not initialize any AMD IOMMU found in 300 the system 301 force_isolation - Force device isolation for all 302 devices. The IOMMU driver is not 303 allowed anymore to lift isolation 304 requirements as needed. This option 305 does not override iommu=pt 306 307 amd_iommu_dump= [HW,X86-64] 308 Enable AMD IOMMU driver option to dump the ACPI table 309 for AMD IOMMU. With this option enabled, AMD IOMMU 310 driver will print ACPI tables for AMD IOMMU during 311 IOMMU initialization. 312 313 amd_iommu_intr= [HW,X86-64] 314 Specifies one of the following AMD IOMMU interrupt 315 remapping modes: 316 legacy - Use legacy interrupt remapping mode. 317 vapic - Use virtual APIC mode, which allows IOMMU 318 to inject interrupts directly into guest. 319 This mode requires kvm-amd.avic=1. 320 (Default when IOMMU HW support is present.) 321 322 amijoy.map= [HW,JOY] Amiga joystick support 323 Map of devices attached to JOY0DAT and JOY1DAT 324 Format: <a>,<b> 325 See also Documentation/input/joydev/joystick.rst 326 327 analog.map= [HW,JOY] Analog joystick and gamepad support 328 Specifies type or capabilities of an analog joystick 329 connected to one of 16 gameports 330 Format: <type1>,<type2>,..<type16> 331 332 apc= [HW,SPARC] 333 Power management functions (SPARCstation-4/5 + deriv.) 334 Format: noidle 335 Disable APC CPU standby support. SPARCstation-Fox does 336 not play well with APC CPU idle - disable it if you have 337 APC and your system crashes randomly. 338 339 apic= [APIC,X86] Advanced Programmable Interrupt Controller 340 Change the output verbosity while booting 341 Format: { quiet (default) | verbose | debug } 342 Change the amount of debugging information output 343 when initialising the APIC and IO-APIC components. 344 For X86-32, this can also be used to specify an APIC 345 driver name. 346 Format: apic=driver_name 347 Examples: apic=bigsmp 348 349 apic_extnmi= [APIC,X86] External NMI delivery setting 350 Format: { bsp (default) | all | none } 351 bsp: External NMI is delivered only to CPU 0 352 all: External NMIs are broadcast to all CPUs as a 353 backup of CPU 0 354 none: External NMI is masked for all CPUs. This is 355 useful so that a dump capture kernel won't be 356 shot down by NMI 357 358 autoconf= [IPV6] 359 See Documentation/networking/ipv6.txt. 360 361 show_lapic= [APIC,X86] Advanced Programmable Interrupt Controller 362 Limit apic dumping. The parameter defines the maximal 363 number of local apics being dumped. Also it is possible 364 to set it to "all" by meaning -- no limit here. 365 Format: { 1 (default) | 2 | ... | all }. 366 The parameter valid if only apic=debug or 367 apic=verbose is specified. 368 Example: apic=debug show_lapic=all 369 370 apm= [APM] Advanced Power Management 371 See header of arch/x86/kernel/apm_32.c. 372 373 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards 374 Format: <io>,<irq>,<nodeID> 375 376 ataflop= [HW,M68k] 377 378 atarimouse= [HW,MOUSE] Atari Mouse 379 380 atkbd.extra= [HW] Enable extra LEDs and keys on IBM RapidAccess, 381 EzKey and similar keyboards 382 383 atkbd.reset= [HW] Reset keyboard during initialization 384 385 atkbd.set= [HW] Select keyboard code set 386 Format: <int> (2 = AT (default), 3 = PS/2) 387 388 atkbd.scroll= [HW] Enable scroll wheel on MS Office and similar 389 keyboards 390 391 atkbd.softraw= [HW] Choose between synthetic and real raw mode 392 Format: <bool> (0 = real, 1 = synthetic (default)) 393 394 atkbd.softrepeat= [HW] 395 Use software keyboard repeat 396 397 audit= [KNL] Enable the audit sub-system 398 Format: { "0" | "1" | "off" | "on" } 399 0 | off - kernel audit is disabled and can not be 400 enabled until the next reboot 401 unset - kernel audit is initialized but disabled and 402 will be fully enabled by the userspace auditd. 403 1 | on - kernel audit is initialized and partially 404 enabled, storing at most audit_backlog_limit 405 messages in RAM until it is fully enabled by the 406 userspace auditd. 407 Default: unset 408 409 audit_backlog_limit= [KNL] Set the audit queue size limit. 410 Format: <int> (must be >=0) 411 Default: 64 412 413 bau= [X86_UV] Enable the BAU on SGI UV. The default 414 behavior is to disable the BAU (i.e. bau=0). 415 Format: { "0" | "1" } 416 0 - Disable the BAU. 417 1 - Enable the BAU. 418 unset - Disable the BAU. 419 420 baycom_epp= [HW,AX25] 421 Format: <io>,<mode> 422 423 baycom_par= [HW,AX25] BayCom Parallel Port AX.25 Modem 424 Format: <io>,<mode> 425 See header of drivers/net/hamradio/baycom_par.c. 426 427 baycom_ser_fdx= [HW,AX25] 428 BayCom Serial Port AX.25 Modem (Full Duplex Mode) 429 Format: <io>,<irq>,<mode>[,<baud>] 430 See header of drivers/net/hamradio/baycom_ser_fdx.c. 431 432 baycom_ser_hdx= [HW,AX25] 433 BayCom Serial Port AX.25 Modem (Half Duplex Mode) 434 Format: <io>,<irq>,<mode> 435 See header of drivers/net/hamradio/baycom_ser_hdx.c. 436 437 blkdevparts= Manual partition parsing of block device(s) for 438 embedded devices based on command line input. 439 See Documentation/block/cmdline-partition.rst 440 441 boot_delay= Milliseconds to delay each printk during boot. 442 Values larger than 10 seconds (10000) are changed to 443 no delay (0). 444 Format: integer 445 446 bootconfig [KNL] 447 Extended command line options can be added to an initrd 448 and this will cause the kernel to look for it. 449 450 See Documentation/admin-guide/bootconfig.rst 451 452 bert_disable [ACPI] 453 Disable BERT OS support on buggy BIOSes. 454 455 bgrt_disable [ACPI][X86] 456 Disable BGRT to avoid flickering OEM logo. 457 458 bttv.card= [HW,V4L] bttv (bt848 + bt878 based grabber cards) 459 bttv.radio= Most important insmod options are available as 460 kernel args too. 461 bttv.pll= See Documentation/media/v4l-drivers/bttv.rst 462 bttv.tuner= 463 464 bulk_remove=off [PPC] This parameter disables the use of the pSeries 465 firmware feature for flushing multiple hpte entries 466 at a time. 467 468 c101= [NET] Moxa C101 synchronous serial card 469 470 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection. 471 Sometimes CPU hardware bugs make them report the cache 472 size incorrectly. The kernel will attempt work arounds 473 to fix known problems, but for some CPUs it is not 474 possible to determine what the correct size should be. 475 This option provides an override for these situations. 476 477 carrier_timeout= 478 [NET] Specifies amount of time (in seconds) that 479 the kernel should wait for a network carrier. By default 480 it waits 120 seconds. 481 482 ca_keys= [KEYS] This parameter identifies a specific key(s) on 483 the system trusted keyring to be used for certificate 484 trust validation. 485 format: { id:<keyid> | builtin } 486 487 cca= [MIPS] Override the kernel pages' cache coherency 488 algorithm. Accepted values range from 0 to 7 489 inclusive. See arch/mips/include/asm/pgtable-bits.h 490 for platform specific values (SB1, Loongson3 and 491 others). 492 493 ccw_timeout_log [S390] 494 See Documentation/s390/common_io.rst for details. 495 496 cgroup_disable= [KNL] Disable a particular controller 497 Format: {name of the controller(s) to disable} 498 The effects of cgroup_disable=foo are: 499 - foo isn't auto-mounted if you mount all cgroups in 500 a single hierarchy 501 - foo isn't visible as an individually mountable 502 subsystem 503 {Currently only "memory" controller deal with this and 504 cut the overhead, others just disable the usage. So 505 only cgroup_disable=memory is actually worthy} 506 507 cgroup_no_v1= [KNL] Disable cgroup controllers and named hierarchies in v1 508 Format: { { controller | "all" | "named" } 509 [,{ controller | "all" | "named" }...] } 510 Like cgroup_disable, but only applies to cgroup v1; 511 the blacklisted controllers remain available in cgroup2. 512 "all" blacklists all controllers and "named" disables 513 named mounts. Specifying both "all" and "named" disables 514 all v1 hierarchies. 515 516 cgroup.memory= [KNL] Pass options to the cgroup memory controller. 517 Format: <string> 518 nosocket -- Disable socket memory accounting. 519 nokmem -- Disable kernel memory accounting. 520 521 checkreqprot [SELINUX] Set initial checkreqprot flag value. 522 Format: { "0" | "1" } 523 See security/selinux/Kconfig help text. 524 0 -- check protection applied by kernel (includes 525 any implied execute protection). 526 1 -- check protection requested by application. 527 Default value is set via a kernel config option. 528 Value can be changed at runtime via 529 /sys/fs/selinux/checkreqprot. 530 Setting checkreqprot to 1 is deprecated. 531 532 cio_ignore= [S390] 533 See Documentation/s390/common_io.rst for details. 534 clk_ignore_unused 535 [CLK] 536 Prevents the clock framework from automatically gating 537 clocks that have not been explicitly enabled by a Linux 538 device driver but are enabled in hardware at reset or 539 by the bootloader/firmware. Note that this does not 540 force such clocks to be always-on nor does it reserve 541 those clocks in any way. This parameter is useful for 542 debug and development, but should not be needed on a 543 platform with proper driver support. For more 544 information, see Documentation/driver-api/clk.rst. 545 546 clock= [BUGS=X86-32, HW] gettimeofday clocksource override. 547 [Deprecated] 548 Forces specified clocksource (if available) to be used 549 when calculating gettimeofday(). If specified 550 clocksource is not available, it defaults to PIT. 551 Format: { pit | tsc | cyclone | pmtmr } 552 553 clocksource= Override the default clocksource 554 Format: <string> 555 Override the default clocksource and use the clocksource 556 with the name specified. 557 Some clocksource names to choose from, depending on 558 the platform: 559 [all] jiffies (this is the base, fallback clocksource) 560 [ACPI] acpi_pm 561 [ARM] imx_timer1,OSTS,netx_timer,mpu_timer2, 562 pxa_timer,timer3,32k_counter,timer0_1 563 [X86-32] pit,hpet,tsc; 564 scx200_hrt on Geode; cyclone on IBM x440 565 [MIPS] MIPS 566 [PARISC] cr16 567 [S390] tod 568 [SH] SuperH 569 [SPARC64] tick 570 [X86-64] hpet,tsc 571 572 clocksource.arm_arch_timer.evtstrm= 573 [ARM,ARM64] 574 Format: <bool> 575 Enable/disable the eventstream feature of the ARM 576 architected timer so that code using WFE-based polling 577 loops can be debugged more effectively on production 578 systems. 579 580 clearcpuid=BITNUM [X86] 581 Disable CPUID feature X for the kernel. See 582 arch/x86/include/asm/cpufeatures.h for the valid bit 583 numbers. Note the Linux specific bits are not necessarily 584 stable over kernel options, but the vendor specific 585 ones should be. 586 Also note that user programs calling CPUID directly 587 or using the feature without checking anything 588 will still see it. This just prevents it from 589 being used by the kernel or shown in /proc/cpuinfo. 590 Also note the kernel might malfunction if you disable 591 some critical bits. 592 593 cma=nn[MG]@[start[MG][-end[MG]]] 594 [ARM,X86,KNL] 595 Sets the size of kernel global memory area for 596 contiguous memory allocations and optionally the 597 placement constraint by the physical address range of 598 memory allocations. A value of 0 disables CMA 599 altogether. For more information, see 600 include/linux/dma-contiguous.h 601 602 cmo_free_hint= [PPC] Format: { yes | no } 603 Specify whether pages are marked as being inactive 604 when they are freed. This is used in CMO environments 605 to determine OS memory pressure for page stealing by 606 a hypervisor. 607 Default: yes 608 609 coherent_pool=nn[KMG] [ARM,KNL] 610 Sets the size of memory pool for coherent, atomic dma 611 allocations, by default set to 256K. 612 613 com20020= [HW,NET] ARCnet - COM20020 chipset 614 Format: 615 <io>[,<irq>[,<nodeID>[,<backplane>[,<ckp>[,<timeout>]]]]] 616 617 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers) 618 Format: <io>[,<irq>] 619 620 com90xx= [HW,NET] 621 ARCnet - COM90xx chipset (memory-mapped buffers) 622 Format: <io>[,<irq>[,<memstart>]] 623 624 condev= [HW,S390] console device 625 conmode= 626 627 console= [KNL] Output console device and options. 628 629 tty<n> Use the virtual console device <n>. 630 631 ttyS<n>[,options] 632 ttyUSB0[,options] 633 Use the specified serial port. The options are of 634 the form "bbbbpnf", where "bbbb" is the baud rate, 635 "p" is parity ("n", "o", or "e"), "n" is number of 636 bits, and "f" is flow control ("r" for RTS or 637 omit it). Default is "9600n8". 638 639 See Documentation/admin-guide/serial-console.rst for more 640 information. See 641 Documentation/networking/netconsole.txt for an 642 alternative. 643 644 uart[8250],io,<addr>[,options] 645 uart[8250],mmio,<addr>[,options] 646 uart[8250],mmio16,<addr>[,options] 647 uart[8250],mmio32,<addr>[,options] 648 uart[8250],0x<addr>[,options] 649 Start an early, polled-mode console on the 8250/16550 650 UART at the specified I/O port or MMIO address, 651 switching to the matching ttyS device later. 652 MMIO inter-register address stride is either 8-bit 653 (mmio), 16-bit (mmio16), or 32-bit (mmio32). 654 If none of [io|mmio|mmio16|mmio32], <addr> is assumed 655 to be equivalent to 'mmio'. 'options' are specified in 656 the same format described for ttyS above; if unspecified, 657 the h/w is not re-initialized. 658 659 hvc<n> Use the hypervisor console device <n>. This is for 660 both Xen and PowerPC hypervisors. 661 662 If the device connected to the port is not a TTY but a braille 663 device, prepend "brl," before the device type, for instance 664 console=brl,ttyS0 665 For now, only VisioBraille is supported. 666 667 console_msg_format= 668 [KNL] Change console messages format 669 default 670 By default we print messages on consoles in 671 "[time stamp] text\n" format (time stamp may not be 672 printed, depending on CONFIG_PRINTK_TIME or 673 `printk_time' param). 674 syslog 675 Switch to syslog format: "<%u>[time stamp] text\n" 676 IOW, each message will have a facility and loglevel 677 prefix. The format is similar to one used by syslog() 678 syscall, or to executing "dmesg -S --raw" or to reading 679 from /proc/kmsg. 680 681 consoleblank= [KNL] The console blank (screen saver) timeout in 682 seconds. A value of 0 disables the blank timer. 683 Defaults to 0. 684 685 coredump_filter= 686 [KNL] Change the default value for 687 /proc/<pid>/coredump_filter. 688 See also Documentation/filesystems/proc.txt. 689 690 coresight_cpu_debug.enable 691 [ARM,ARM64] 692 Format: <bool> 693 Enable/disable the CPU sampling based debugging. 694 0: default value, disable debugging 695 1: enable debugging at boot time 696 697 cpuidle.off=1 [CPU_IDLE] 698 disable the cpuidle sub-system 699 700 cpuidle.governor= 701 [CPU_IDLE] Name of the cpuidle governor to use. 702 703 cpufreq.off=1 [CPU_FREQ] 704 disable the cpufreq sub-system 705 706 cpu_init_udelay=N 707 [X86] Delay for N microsec between assert and de-assert 708 of APIC INIT to start processors. This delay occurs 709 on every CPU online, such as boot, and resume from suspend. 710 Default: 10000 711 712 cpcihp_generic= [HW,PCI] Generic port I/O CompactPCI driver 713 Format: 714 <first_slot>,<last_slot>,<port>,<enum_bit>[,<debug>] 715 716 crashkernel=size[KMG][@offset[KMG]] 717 [KNL] Using kexec, Linux can switch to a 'crash kernel' 718 upon panic. This parameter reserves the physical 719 memory region [offset, offset + size] for that kernel 720 image. If '@offset' is omitted, then a suitable offset 721 is selected automatically. 722 [KNL, x86_64] select a region under 4G first, and 723 fall back to reserve region above 4G when '@offset' 724 hasn't been specified. 725 See Documentation/admin-guide/kdump/kdump.rst for further details. 726 727 crashkernel=range1:size1[,range2:size2,...][@offset] 728 [KNL] Same as above, but depends on the memory 729 in the running system. The syntax of range is 730 start-[end] where start and end are both 731 a memory unit (amount[KMG]). See also 732 Documentation/admin-guide/kdump/kdump.rst for an example. 733 734 crashkernel=size[KMG],high 735 [KNL, x86_64] range could be above 4G. Allow kernel 736 to allocate physical memory region from top, so could 737 be above 4G if system have more than 4G ram installed. 738 Otherwise memory region will be allocated below 4G, if 739 available. 740 It will be ignored if crashkernel=X is specified. 741 crashkernel=size[KMG],low 742 [KNL, x86_64] range under 4G. When crashkernel=X,high 743 is passed, kernel could allocate physical memory region 744 above 4G, that cause second kernel crash on system 745 that require some amount of low memory, e.g. swiotlb 746 requires at least 64M+32K low memory, also enough extra 747 low memory is needed to make sure DMA buffers for 32-bit 748 devices won't run out. Kernel would try to allocate at 749 at least 256M below 4G automatically. 750 This one let user to specify own low range under 4G 751 for second kernel instead. 752 0: to disable low allocation. 753 It will be ignored when crashkernel=X,high is not used 754 or memory reserved is below 4G. 755 756 cryptomgr.notests 757 [KNL] Disable crypto self-tests 758 759 cs89x0_dma= [HW,NET] 760 Format: <dma> 761 762 cs89x0_media= [HW,NET] 763 Format: { rj45 | aui | bnc } 764 765 dasd= [HW,NET] 766 See header of drivers/s390/block/dasd_devmap.c. 767 768 db9.dev[2|3]= [HW,JOY] Multisystem joystick support via parallel port 769 (one device per port) 770 Format: <port#>,<type> 771 See also Documentation/input/devices/joystick-parport.rst 772 773 ddebug_query= [KNL,DYNAMIC_DEBUG] Enable debug messages at early boot 774 time. See 775 Documentation/admin-guide/dynamic-debug-howto.rst for 776 details. Deprecated, see dyndbg. 777 778 debug [KNL] Enable kernel debugging (events log level). 779 780 debug_boot_weak_hash 781 [KNL] Enable printing [hashed] pointers early in the 782 boot sequence. If enabled, we use a weak hash instead 783 of siphash to hash pointers. Use this option if you are 784 seeing instances of '(___ptrval___)') and need to see a 785 value (hashed pointer) instead. Cryptographically 786 insecure, please do not use on production kernels. 787 788 debug_locks_verbose= 789 [KNL] verbose self-tests 790 Format=<0|1> 791 Print debugging info while doing the locking API 792 self-tests. 793 We default to 0 (no extra messages), setting it to 794 1 will print _a lot_ more information - normally 795 only useful to kernel developers. 796 797 debug_objects [KNL] Enable object debugging 798 799 no_debug_objects 800 [KNL] Disable object debugging 801 802 debug_guardpage_minorder= 803 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this 804 parameter allows control of the order of pages that will 805 be intentionally kept free (and hence protected) by the 806 buddy allocator. Bigger value increase the probability 807 of catching random memory corruption, but reduce the 808 amount of memory for normal system use. The maximum 809 possible value is MAX_ORDER/2. Setting this parameter 810 to 1 or 2 should be enough to identify most random 811 memory corruption problems caused by bugs in kernel or 812 driver code when a CPU writes to (or reads from) a 813 random memory location. Note that there exists a class 814 of memory corruptions problems caused by buggy H/W or 815 F/W or by drivers badly programing DMA (basically when 816 memory is written at bus level and the CPU MMU is 817 bypassed) which are not detectable by 818 CONFIG_DEBUG_PAGEALLOC, hence this option will not help 819 tracking down these problems. 820 821 debug_pagealloc= 822 [KNL] When CONFIG_DEBUG_PAGEALLOC is set, this parameter 823 enables the feature at boot time. By default, it is 824 disabled and the system will work mostly the same as a 825 kernel built without CONFIG_DEBUG_PAGEALLOC. 826 Note: to get most of debug_pagealloc error reports, it's 827 useful to also enable the page_owner functionality. 828 on: enable the feature 829 830 debugpat [X86] Enable PAT debugging 831 832 decnet.addr= [HW,NET] 833 Format: <area>[,<node>] 834 See also Documentation/networking/decnet.txt. 835 836 default_hugepagesz= 837 [same as hugepagesz=] The size of the default 838 HugeTLB page size. This is the size represented by 839 the legacy /proc/ hugepages APIs, used for SHM, and 840 default size when mounting hugetlbfs filesystems. 841 Defaults to the default architecture's huge page size 842 if not specified. 843 844 deferred_probe_timeout= 845 [KNL] Debugging option to set a timeout in seconds for 846 deferred probe to give up waiting on dependencies to 847 probe. Only specific dependencies (subsystems or 848 drivers) that have opted in will be ignored. A timeout of 0 849 will timeout at the end of initcalls. This option will also 850 dump out devices still on the deferred probe list after 851 retrying. 852 853 dfltcc= [HW,S390] 854 Format: { on | off | def_only | inf_only | always } 855 on: s390 zlib hardware support for compression on 856 level 1 and decompression (default) 857 off: No s390 zlib hardware support 858 def_only: s390 zlib hardware support for deflate 859 only (compression on level 1) 860 inf_only: s390 zlib hardware support for inflate 861 only (decompression) 862 always: Same as 'on' but ignores the selected compression 863 level always using hardware support (used for debugging) 864 865 dhash_entries= [KNL] 866 Set number of hash buckets for dentry cache. 867 868 disable_1tb_segments [PPC] 869 Disables the use of 1TB hash page table segments. This 870 causes the kernel to fall back to 256MB segments which 871 can be useful when debugging issues that require an SLB 872 miss to occur. 873 874 disable= [IPV6] 875 See Documentation/networking/ipv6.txt. 876 877 hardened_usercopy= 878 [KNL] Under CONFIG_HARDENED_USERCOPY, whether 879 hardening is enabled for this boot. Hardened 880 usercopy checking is used to protect the kernel 881 from reading or writing beyond known memory 882 allocation boundaries as a proactive defense 883 against bounds-checking flaws in the kernel's 884 copy_to_user()/copy_from_user() interface. 885 on Perform hardened usercopy checks (default). 886 off Disable hardened usercopy checks. 887 888 disable_radix [PPC] 889 Disable RADIX MMU mode on POWER9 890 891 disable_tlbie [PPC] 892 Disable TLBIE instruction. Currently does not work 893 with KVM, with HASH MMU, or with coherent accelerators. 894 895 disable_cpu_apicid= [X86,APIC,SMP] 896 Format: <int> 897 The number of initial APIC ID for the 898 corresponding CPU to be disabled at boot, 899 mostly used for the kdump 2nd kernel to 900 disable BSP to wake up multiple CPUs without 901 causing system reset or hang due to sending 902 INIT from AP to BSP. 903 904 perf_v4_pmi= [X86,INTEL] 905 Format: <bool> 906 Disable Intel PMU counter freezing feature. 907 The feature only exists starting from 908 Arch Perfmon v4 (Skylake and newer). 909 910 disable_ddw [PPC/PSERIES] 911 Disable Dynamic DMA Window support. Use this if 912 to workaround buggy firmware. 913 914 disable_ipv6= [IPV6] 915 See Documentation/networking/ipv6.txt. 916 917 disable_mtrr_cleanup [X86] 918 The kernel tries to adjust MTRR layout from continuous 919 to discrete, to make X server driver able to add WB 920 entry later. This parameter disables that. 921 922 disable_mtrr_trim [X86, Intel and AMD only] 923 By default the kernel will trim any uncacheable 924 memory out of your available memory pool based on 925 MTRR settings. This parameter disables that behavior, 926 possibly causing your machine to run very slowly. 927 928 disable_timer_pin_1 [X86] 929 Disable PIN 1 of APIC timer 930 Can be useful to work around chipset bugs. 931 932 dis_ucode_ldr [X86] Disable the microcode loader. 933 934 dma_debug=off If the kernel is compiled with DMA_API_DEBUG support, 935 this option disables the debugging code at boot. 936 937 dma_debug_entries=<number> 938 This option allows to tune the number of preallocated 939 entries for DMA-API debugging code. One entry is 940 required per DMA-API allocation. Use this if the 941 DMA-API debugging code disables itself because the 942 architectural default is too low. 943 944 dma_debug_driver=<driver_name> 945 With this option the DMA-API debugging driver 946 filter feature can be enabled at boot time. Just 947 pass the driver to filter for as the parameter. 948 The filter can be disabled or changed to another 949 driver later using sysfs. 950 951 driver_async_probe= [KNL] 952 List of driver names to be probed asynchronously. 953 Format: <driver_name1>,<driver_name2>... 954 955 drm.edid_firmware=[<connector>:]<file>[,[<connector>:]<file>] 956 Broken monitors, graphic adapters, KVMs and EDIDless 957 panels may send no or incorrect EDID data sets. 958 This parameter allows to specify an EDID data sets 959 in the /lib/firmware directory that are used instead. 960 Generic built-in EDID data sets are used, if one of 961 edid/1024x768.bin, edid/1280x1024.bin, 962 edid/1680x1050.bin, or edid/1920x1080.bin is given 963 and no file with the same name exists. Details and 964 instructions how to build your own EDID data are 965 available in Documentation/driver-api/edid.rst. An EDID 966 data set will only be used for a particular connector, 967 if its name and a colon are prepended to the EDID 968 name. Each connector may use a unique EDID data 969 set by separating the files with a comma. An EDID 970 data set with no connector name will be used for 971 any connectors not explicitly specified. 972 973 dscc4.setup= [NET] 974 975 dt_cpu_ftrs= [PPC] 976 Format: {"off" | "known"} 977 Control how the dt_cpu_ftrs device-tree binding is 978 used for CPU feature discovery and setup (if it 979 exists). 980 off: Do not use it, fall back to legacy cpu table. 981 known: Do not pass through unknown features to guests 982 or userspace, only those that the kernel is aware of. 983 984 dump_apple_properties [X86] 985 Dump name and content of EFI device properties on 986 x86 Macs. Useful for driver authors to determine 987 what data is available or for reverse-engineering. 988 989 dyndbg[="val"] [KNL,DYNAMIC_DEBUG] 990 module.dyndbg[="val"] 991 Enable debug messages at boot time. See 992 Documentation/admin-guide/dynamic-debug-howto.rst 993 for details. 994 995 nompx [X86] Disables Intel Memory Protection Extensions. 996 See Documentation/x86/intel_mpx.rst for more 997 information about the feature. 998 999 nopku [X86] Disable Memory Protection Keys CPU feature found 1000 in some Intel CPUs. 1001 1002 module.async_probe [KNL] 1003 Enable asynchronous probe on this module. 1004 1005 early_ioremap_debug [KNL] 1006 Enable debug messages in early_ioremap support. This 1007 is useful for tracking down temporary early mappings 1008 which are not unmapped. 1009 1010 earlycon= [KNL] Output early console device and options. 1011 1012 When used with no options, the early console is 1013 determined by stdout-path property in device tree's 1014 chosen node or the ACPI SPCR table if supported by 1015 the platform. 1016 1017 cdns,<addr>[,options] 1018 Start an early, polled-mode console on a Cadence 1019 (xuartps) serial port at the specified address. Only 1020 supported option is baud rate. If baud rate is not 1021 specified, the serial port must already be setup and 1022 configured. 1023 1024 uart[8250],io,<addr>[,options] 1025 uart[8250],mmio,<addr>[,options] 1026 uart[8250],mmio32,<addr>[,options] 1027 uart[8250],mmio32be,<addr>[,options] 1028 uart[8250],0x<addr>[,options] 1029 Start an early, polled-mode console on the 8250/16550 1030 UART at the specified I/O port or MMIO address. 1031 MMIO inter-register address stride is either 8-bit 1032 (mmio) or 32-bit (mmio32 or mmio32be). 1033 If none of [io|mmio|mmio32|mmio32be], <addr> is assumed 1034 to be equivalent to 'mmio'. 'options' are specified 1035 in the same format described for "console=ttyS<n>"; if 1036 unspecified, the h/w is not initialized. 1037 1038 pl011,<addr> 1039 pl011,mmio32,<addr> 1040 Start an early, polled-mode console on a pl011 serial 1041 port at the specified address. The pl011 serial port 1042 must already be setup and configured. Options are not 1043 yet supported. If 'mmio32' is specified, then only 1044 the driver will use only 32-bit accessors to read/write 1045 the device registers. 1046 1047 meson,<addr> 1048 Start an early, polled-mode console on a meson serial 1049 port at the specified address. The serial port must 1050 already be setup and configured. Options are not yet 1051 supported. 1052 1053 msm_serial,<addr> 1054 Start an early, polled-mode console on an msm serial 1055 port at the specified address. The serial port 1056 must already be setup and configured. Options are not 1057 yet supported. 1058 1059 msm_serial_dm,<addr> 1060 Start an early, polled-mode console on an msm serial 1061 dm port at the specified address. The serial port 1062 must already be setup and configured. Options are not 1063 yet supported. 1064 1065 owl,<addr> 1066 Start an early, polled-mode console on a serial port 1067 of an Actions Semi SoC, such as S500 or S900, at the 1068 specified address. The serial port must already be 1069 setup and configured. Options are not yet supported. 1070 1071 rda,<addr> 1072 Start an early, polled-mode console on a serial port 1073 of an RDA Micro SoC, such as RDA8810PL, at the 1074 specified address. The serial port must already be 1075 setup and configured. Options are not yet supported. 1076 1077 sbi 1078 Use RISC-V SBI (Supervisor Binary Interface) for early 1079 console. 1080 1081 smh Use ARM semihosting calls for early console. 1082 1083 s3c2410,<addr> 1084 s3c2412,<addr> 1085 s3c2440,<addr> 1086 s3c6400,<addr> 1087 s5pv210,<addr> 1088 exynos4210,<addr> 1089 Use early console provided by serial driver available 1090 on Samsung SoCs, requires selecting proper type and 1091 a correct base address of the selected UART port. The 1092 serial port must already be setup and configured. 1093 Options are not yet supported. 1094 1095 lantiq,<addr> 1096 Start an early, polled-mode console on a lantiq serial 1097 (lqasc) port at the specified address. The serial port 1098 must already be setup and configured. Options are not 1099 yet supported. 1100 1101 lpuart,<addr> 1102 lpuart32,<addr> 1103 Use early console provided by Freescale LP UART driver 1104 found on Freescale Vybrid and QorIQ LS1021A processors. 1105 A valid base address must be provided, and the serial 1106 port must already be setup and configured. 1107 1108 ec_imx21,<addr> 1109 ec_imx6q,<addr> 1110 Start an early, polled-mode, output-only console on the 1111 Freescale i.MX UART at the specified address. The UART 1112 must already be setup and configured. 1113 1114 ar3700_uart,<addr> 1115 Start an early, polled-mode console on the 1116 Armada 3700 serial port at the specified 1117 address. The serial port must already be setup 1118 and configured. Options are not yet supported. 1119 1120 qcom_geni,<addr> 1121 Start an early, polled-mode console on a Qualcomm 1122 Generic Interface (GENI) based serial port at the 1123 specified address. The serial port must already be 1124 setup and configured. Options are not yet supported. 1125 1126 efifb,[options] 1127 Start an early, unaccelerated console on the EFI 1128 memory mapped framebuffer (if available). On cache 1129 coherent non-x86 systems that use system memory for 1130 the framebuffer, pass the 'ram' option so that it is 1131 mapped with the correct attributes. 1132 1133 linflex,<addr> 1134 Use early console provided by Freescale LINFlexD UART 1135 serial driver for NXP S32V234 SoCs. A valid base 1136 address must be provided, and the serial port must 1137 already be setup and configured. 1138 1139 earlyprintk= [X86,SH,ARM,M68k,S390] 1140 earlyprintk=vga 1141 earlyprintk=sclp 1142 earlyprintk=xen 1143 earlyprintk=serial[,ttySn[,baudrate]] 1144 earlyprintk=serial[,0x...[,baudrate]] 1145 earlyprintk=ttySn[,baudrate] 1146 earlyprintk=dbgp[debugController#] 1147 earlyprintk=pciserial[,force],bus:device.function[,baudrate] 1148 earlyprintk=xdbc[xhciController#] 1149 1150 earlyprintk is useful when the kernel crashes before 1151 the normal console is initialized. It is not enabled by 1152 default because it has some cosmetic problems. 1153 1154 Append ",keep" to not disable it when the real console 1155 takes over. 1156 1157 Only one of vga, efi, serial, or usb debug port can 1158 be used at a time. 1159 1160 Currently only ttyS0 and ttyS1 may be specified by 1161 name. Other I/O ports may be explicitly specified 1162 on some architectures (x86 and arm at least) by 1163 replacing ttySn with an I/O port address, like this: 1164 earlyprintk=serial,0x1008,115200 1165 You can find the port for a given device in 1166 /proc/tty/driver/serial: 1167 2: uart:ST16650V2 port:00001008 irq:18 ... 1168 1169 Interaction with the standard serial driver is not 1170 very good. 1171 1172 The VGA and EFI output is eventually overwritten by 1173 the real console. 1174 1175 The xen output can only be used by Xen PV guests. 1176 1177 The sclp output can only be used on s390. 1178 1179 The optional "force" to "pciserial" enables use of a 1180 PCI device even when its classcode is not of the 1181 UART class. 1182 1183 edac_report= [HW,EDAC] Control how to report EDAC event 1184 Format: {"on" | "off" | "force"} 1185 on: enable EDAC to report H/W event. May be overridden 1186 by other higher priority error reporting module. 1187 off: disable H/W event reporting through EDAC. 1188 force: enforce the use of EDAC to report H/W event. 1189 default: on. 1190 1191 ekgdboc= [X86,KGDB] Allow early kernel console debugging 1192 ekgdboc=kbd 1193 1194 This is designed to be used in conjunction with 1195 the boot argument: earlyprintk=vga 1196 1197 edd= [EDD] 1198 Format: {"off" | "on" | "skip[mbr]"} 1199 1200 efi= [EFI] 1201 Format: { "old_map", "nochunk", "noruntime", "debug", 1202 "nosoftreserve", "disable_early_pci_dma", 1203 "no_disable_early_pci_dma" } 1204 old_map [X86-64]: switch to the old ioremap-based EFI 1205 runtime services mapping. [Needs CONFIG_X86_UV=y] 1206 nochunk: disable reading files in "chunks" in the EFI 1207 boot stub, as chunking can cause problems with some 1208 firmware implementations. 1209 noruntime : disable EFI runtime services support 1210 debug: enable misc debug output 1211 nosoftreserve: The EFI_MEMORY_SP (Specific Purpose) 1212 attribute may cause the kernel to reserve the 1213 memory range for a memory mapping driver to 1214 claim. Specify efi=nosoftreserve to disable this 1215 reservation and treat the memory by its base type 1216 (i.e. EFI_CONVENTIONAL_MEMORY / "System RAM"). 1217 disable_early_pci_dma: Disable the busmaster bit on all 1218 PCI bridges while in the EFI boot stub 1219 no_disable_early_pci_dma: Leave the busmaster bit set 1220 on all PCI bridges while in the EFI boot stub 1221 1222 efi_no_storage_paranoia [EFI; X86] 1223 Using this parameter you can use more than 50% of 1224 your efi variable storage. Use this parameter only if 1225 you are really sure that your UEFI does sane gc and 1226 fulfills the spec otherwise your board may brick. 1227 1228 efi_fake_mem= nn[KMG]@ss[KMG]:aa[,nn[KMG]@ss[KMG]:aa,..] [EFI; X86] 1229 Add arbitrary attribute to specific memory range by 1230 updating original EFI memory map. 1231 Region of memory which aa attribute is added to is 1232 from ss to ss+nn. 1233 1234 If efi_fake_mem=2G@4G:0x10000,2G@0x10a0000000:0x10000 1235 is specified, EFI_MEMORY_MORE_RELIABLE(0x10000) 1236 attribute is added to range 0x100000000-0x180000000 and 1237 0x10a0000000-0x1120000000. 1238 1239 If efi_fake_mem=8G@9G:0x40000 is specified, the 1240 EFI_MEMORY_SP(0x40000) attribute is added to 1241 range 0x240000000-0x43fffffff. 1242 1243 Using this parameter you can do debugging of EFI memmap 1244 related features. For example, you can do debugging of 1245 Address Range Mirroring feature even if your box 1246 doesn't support it, or mark specific memory as 1247 "soft reserved". 1248 1249 efivar_ssdt= [EFI; X86] Name of an EFI variable that contains an SSDT 1250 that is to be dynamically loaded by Linux. If there are 1251 multiple variables with the same name but with different 1252 vendor GUIDs, all of them will be loaded. See 1253 Documentation/admin-guide/acpi/ssdt-overlays.rst for details. 1254 1255 1256 eisa_irq_edge= [PARISC,HW] 1257 See header of drivers/parisc/eisa.c. 1258 1259 elanfreq= [X86-32] 1260 See comment before function elanfreq_setup() in 1261 arch/x86/kernel/cpu/cpufreq/elanfreq.c. 1262 1263 elfcorehdr=[size[KMG]@]offset[KMG] [IA64,PPC,SH,X86,S390] 1264 Specifies physical address of start of kernel core 1265 image elf header and optionally the size. Generally 1266 kexec loader will pass this option to capture kernel. 1267 See Documentation/admin-guide/kdump/kdump.rst for details. 1268 1269 enable_mtrr_cleanup [X86] 1270 The kernel tries to adjust MTRR layout from continuous 1271 to discrete, to make X server driver able to add WB 1272 entry later. This parameter enables that. 1273 1274 enable_timer_pin_1 [X86] 1275 Enable PIN 1 of APIC timer 1276 Can be useful to work around chipset bugs 1277 (in particular on some ATI chipsets). 1278 The kernel tries to set a reasonable default. 1279 1280 enforcing [SELINUX] Set initial enforcing status. 1281 Format: {"0" | "1"} 1282 See security/selinux/Kconfig help text. 1283 0 -- permissive (log only, no denials). 1284 1 -- enforcing (deny and log). 1285 Default value is 0. 1286 Value can be changed at runtime via 1287 /sys/fs/selinux/enforce. 1288 1289 erst_disable [ACPI] 1290 Disable Error Record Serialization Table (ERST) 1291 support. 1292 1293 ether= [HW,NET] Ethernet cards parameters 1294 This option is obsoleted by the "netdev=" option, which 1295 has equivalent usage. See its documentation for details. 1296 1297 evm= [EVM] 1298 Format: { "fix" } 1299 Permit 'security.evm' to be updated regardless of 1300 current integrity status. 1301 1302 failslab= 1303 fail_page_alloc= 1304 fail_make_request=[KNL] 1305 General fault injection mechanism. 1306 Format: <interval>,<probability>,<space>,<times> 1307 See also Documentation/fault-injection/. 1308 1309 floppy= [HW] 1310 See Documentation/admin-guide/blockdev/floppy.rst. 1311 1312 force_pal_cache_flush 1313 [IA-64] Avoid check_sal_cache_flush which may hang on 1314 buggy SAL_CACHE_FLUSH implementations. Using this 1315 parameter will force ia64_sal_cache_flush to call 1316 ia64_pal_cache_flush instead of SAL_CACHE_FLUSH. 1317 1318 forcepae [X86-32] 1319 Forcefully enable Physical Address Extension (PAE). 1320 Many Pentium M systems disable PAE but may have a 1321 functionally usable PAE implementation. 1322 Warning: use of this parameter will taint the kernel 1323 and may cause unknown problems. 1324 1325 ftrace=[tracer] 1326 [FTRACE] will set and start the specified tracer 1327 as early as possible in order to facilitate early 1328 boot debugging. 1329 1330 ftrace_dump_on_oops[=orig_cpu] 1331 [FTRACE] will dump the trace buffers on oops. 1332 If no parameter is passed, ftrace will dump 1333 buffers of all CPUs, but if you pass orig_cpu, it will 1334 dump only the buffer of the CPU that triggered the 1335 oops. 1336 1337 ftrace_filter=[function-list] 1338 [FTRACE] Limit the functions traced by the function 1339 tracer at boot up. function-list is a comma separated 1340 list of functions. This list can be changed at run 1341 time by the set_ftrace_filter file in the debugfs 1342 tracing directory. 1343 1344 ftrace_notrace=[function-list] 1345 [FTRACE] Do not trace the functions specified in 1346 function-list. This list can be changed at run time 1347 by the set_ftrace_notrace file in the debugfs 1348 tracing directory. 1349 1350 ftrace_graph_filter=[function-list] 1351 [FTRACE] Limit the top level callers functions traced 1352 by the function graph tracer at boot up. 1353 function-list is a comma separated list of functions 1354 that can be changed at run time by the 1355 set_graph_function file in the debugfs tracing directory. 1356 1357 ftrace_graph_notrace=[function-list] 1358 [FTRACE] Do not trace from the functions specified in 1359 function-list. This list is a comma separated list of 1360 functions that can be changed at run time by the 1361 set_graph_notrace file in the debugfs tracing directory. 1362 1363 ftrace_graph_max_depth=<uint> 1364 [FTRACE] Used with the function graph tracer. This is 1365 the max depth it will trace into a function. This value 1366 can be changed at run time by the max_graph_depth file 1367 in the tracefs tracing directory. default: 0 (no limit) 1368 1369 fw_devlink= [KNL] Create device links between consumer and supplier 1370 devices by scanning the firmware to infer the 1371 consumer/supplier relationships. This feature is 1372 especially useful when drivers are loaded as modules as 1373 it ensures proper ordering of tasks like device probing 1374 (suppliers first, then consumers), supplier boot state 1375 clean up (only after all consumers have probed), 1376 suspend/resume & runtime PM (consumers first, then 1377 suppliers). 1378 Format: { off | permissive | on | rpm } 1379 off -- Don't create device links from firmware info. 1380 permissive -- Create device links from firmware info 1381 but use it only for ordering boot state clean 1382 up (sync_state() calls). 1383 on -- Create device links from firmware info and use it 1384 to enforce probe and suspend/resume ordering. 1385 rpm -- Like "on", but also use to order runtime PM. 1386 1387 gamecon.map[2|3]= 1388 [HW,JOY] Multisystem joystick and NES/SNES/PSX pad 1389 support via parallel port (up to 5 devices per port) 1390 Format: <port#>,<pad1>,<pad2>,<pad3>,<pad4>,<pad5> 1391 See also Documentation/input/devices/joystick-parport.rst 1392 1393 gamma= [HW,DRM] 1394 1395 gart_fix_e820= [X86_64] disable the fix e820 for K8 GART 1396 Format: off | on 1397 default: on 1398 1399 gcov_persist= [GCOV] When non-zero (default), profiling data for 1400 kernel modules is saved and remains accessible via 1401 debugfs, even when the module is unloaded/reloaded. 1402 When zero, profiling data is discarded and associated 1403 debugfs files are removed at module unload time. 1404 1405 goldfish [X86] Enable the goldfish android emulator platform. 1406 Don't use this when you are not running on the 1407 android emulator 1408 1409 gpt [EFI] Forces disk with valid GPT signature but 1410 invalid Protective MBR to be treated as GPT. If the 1411 primary GPT is corrupted, it enables the backup/alternate 1412 GPT to be used instead. 1413 1414 grcan.enable0= [HW] Configuration of physical interface 0. Determines 1415 the "Enable 0" bit of the configuration register. 1416 Format: 0 | 1 1417 Default: 0 1418 grcan.enable1= [HW] Configuration of physical interface 1. Determines 1419 the "Enable 0" bit of the configuration register. 1420 Format: 0 | 1 1421 Default: 0 1422 grcan.select= [HW] Select which physical interface to use. 1423 Format: 0 | 1 1424 Default: 0 1425 grcan.txsize= [HW] Sets the size of the tx buffer. 1426 Format: <unsigned int> such that (txsize & ~0x1fffc0) == 0. 1427 Default: 1024 1428 grcan.rxsize= [HW] Sets the size of the rx buffer. 1429 Format: <unsigned int> such that (rxsize & ~0x1fffc0) == 0. 1430 Default: 1024 1431 1432 gpio-mockup.gpio_mockup_ranges 1433 [HW] Sets the ranges of gpiochip of for this device. 1434 Format: <start1>,<end1>,<start2>,<end2>... 1435 1436 hardlockup_all_cpu_backtrace= 1437 [KNL] Should the hard-lockup detector generate 1438 backtraces on all cpus. 1439 Format: <integer> 1440 1441 hashdist= [KNL,NUMA] Large hashes allocated during boot 1442 are distributed across NUMA nodes. Defaults on 1443 for 64-bit NUMA, off otherwise. 1444 Format: 0 | 1 (for off | on) 1445 1446 hcl= [IA-64] SGI's Hardware Graph compatibility layer 1447 1448 hd= [EIDE] (E)IDE hard drive subsystem geometry 1449 Format: <cyl>,<head>,<sect> 1450 1451 hest_disable [ACPI] 1452 Disable Hardware Error Source Table (HEST) support; 1453 corresponding firmware-first mode error processing 1454 logic will be disabled. 1455 1456 highmem=nn[KMG] [KNL,BOOT] forces the highmem zone to have an exact 1457 size of <nn>. This works even on boxes that have no 1458 highmem otherwise. This also works to reduce highmem 1459 size on bigger boxes. 1460 1461 highres= [KNL] Enable/disable high resolution timer mode. 1462 Valid parameters: "on", "off" 1463 Default: "on" 1464 1465 hlt [BUGS=ARM,SH] 1466 1467 hpet= [X86-32,HPET] option to control HPET usage 1468 Format: { enable (default) | disable | force | 1469 verbose } 1470 disable: disable HPET and use PIT instead 1471 force: allow force enabled of undocumented chips (ICH4, 1472 VIA, nVidia) 1473 verbose: show contents of HPET registers during setup 1474 1475 hpet_mmap= [X86, HPET_MMAP] Allow userspace to mmap HPET 1476 registers. Default set by CONFIG_HPET_MMAP_DEFAULT. 1477 1478 hugepages= [HW,X86-32,IA-64] HugeTLB pages to allocate at boot. 1479 hugepagesz= [HW,IA-64,PPC,X86-64] The size of the HugeTLB pages. 1480 On x86-64 and powerpc, this option can be specified 1481 multiple times interleaved with hugepages= to reserve 1482 huge pages of different sizes. Valid pages sizes on 1483 x86-64 are 2M (when the CPU supports "pse") and 1G 1484 (when the CPU supports the "pdpe1gb" cpuinfo flag). 1485 1486 hung_task_panic= 1487 [KNL] Should the hung task detector generate panics. 1488 Format: <integer> 1489 1490 A nonzero value instructs the kernel to panic when a 1491 hung task is detected. The default value is controlled 1492 by the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time 1493 option. The value selected by this boot parameter can 1494 be changed later by the kernel.hung_task_panic sysctl. 1495 1496 hvc_iucv= [S390] Number of z/VM IUCV hypervisor console (HVC) 1497 terminal devices. Valid values: 0..8 1498 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs. 1499 If specified, z/VM IUCV HVC accepts connections 1500 from listed z/VM user IDs only. 1501 1502 hv_nopvspin [X86,HYPER_V] Disables the paravirt spinlock optimizations 1503 which allow the hypervisor to 'idle' the 1504 guest on lock contention. 1505 1506 keep_bootcon [KNL] 1507 Do not unregister boot console at start. This is only 1508 useful for debugging when something happens in the window 1509 between unregistering the boot console and initializing 1510 the real console. 1511 1512 i2c_bus= [HW] Override the default board specific I2C bus speed 1513 or register an additional I2C bus that is not 1514 registered from board initialization code. 1515 Format: 1516 <bus_id>,<clkrate> 1517 1518 i8042.debug [HW] Toggle i8042 debug mode 1519 i8042.unmask_kbd_data 1520 [HW] Enable printing of interrupt data from the KBD port 1521 (disabled by default, and as a pre-condition 1522 requires that i8042.debug=1 be enabled) 1523 i8042.direct [HW] Put keyboard port into non-translated mode 1524 i8042.dumbkbd [HW] Pretend that controller can only read data from 1525 keyboard and cannot control its state 1526 (Don't attempt to blink the leds) 1527 i8042.noaux [HW] Don't check for auxiliary (== mouse) port 1528 i8042.nokbd [HW] Don't check/create keyboard port 1529 i8042.noloop [HW] Disable the AUX Loopback command while probing 1530 for the AUX port 1531 i8042.nomux [HW] Don't check presence of an active multiplexing 1532 controller 1533 i8042.nopnp [HW] Don't use ACPIPnP / PnPBIOS to discover KBD/AUX 1534 controllers 1535 i8042.notimeout [HW] Ignore timeout condition signalled by controller 1536 i8042.reset [HW] Reset the controller during init, cleanup and 1537 suspend-to-ram transitions, only during s2r 1538 transitions, or never reset 1539 Format: { 1 | Y | y | 0 | N | n } 1540 1, Y, y: always reset controller 1541 0, N, n: don't ever reset controller 1542 Default: only on s2r transitions on x86; most other 1543 architectures force reset to be always executed 1544 i8042.unlock [HW] Unlock (ignore) the keylock 1545 i8042.kbdreset [HW] Reset device connected to KBD port 1546 1547 i810= [HW,DRM] 1548 1549 i8k.ignore_dmi [HW] Continue probing hardware even if DMI data 1550 indicates that the driver is running on unsupported 1551 hardware. 1552 i8k.force [HW] Activate i8k driver even if SMM BIOS signature 1553 does not match list of supported models. 1554 i8k.power_status 1555 [HW] Report power status in /proc/i8k 1556 (disabled by default) 1557 i8k.restricted [HW] Allow controlling fans only if SYS_ADMIN 1558 capability is set. 1559 1560 i915.invert_brightness= 1561 [DRM] Invert the sense of the variable that is used to 1562 set the brightness of the panel backlight. Normally a 1563 brightness value of 0 indicates backlight switched off, 1564 and the maximum of the brightness value sets the backlight 1565 to maximum brightness. If this parameter is set to 0 1566 (default) and the machine requires it, or this parameter 1567 is set to 1, a brightness value of 0 sets the backlight 1568 to maximum brightness, and the maximum of the brightness 1569 value switches the backlight off. 1570 -1 -- never invert brightness 1571 0 -- machine default 1572 1 -- force brightness inversion 1573 1574 icn= [HW,ISDN] 1575 Format: <io>[,<membase>[,<icn_id>[,<icn_id2>]]] 1576 1577 ide-core.nodma= [HW] (E)IDE subsystem 1578 Format: =0.0 to prevent dma on hda, =0.1 hdb =1.0 hdc 1579 .vlb_clock .pci_clock .noflush .nohpa .noprobe .nowerr 1580 .cdrom .chs .ignore_cable are additional options 1581 See Documentation/ide/ide.rst. 1582 1583 ide-generic.probe-mask= [HW] (E)IDE subsystem 1584 Format: <int> 1585 Probe mask for legacy ISA IDE ports. Depending on 1586 platform up to 6 ports are supported, enabled by 1587 setting corresponding bits in the mask to 1. The 1588 default value is 0x0, which has a special meaning. 1589 On systems that have PCI, it triggers scanning the 1590 PCI bus for the first and the second port, which 1591 are then probed. On systems without PCI the value 1592 of 0x0 enables probing the two first ports as if it 1593 was 0x3. 1594 1595 ide-pci-generic.all-generic-ide [HW] (E)IDE subsystem 1596 Claim all unknown PCI IDE storage controllers. 1597 1598 idle= [X86] 1599 Format: idle=poll, idle=halt, idle=nomwait 1600 Poll forces a polling idle loop that can slightly 1601 improve the performance of waking up a idle CPU, but 1602 will use a lot of power and make the system run hot. 1603 Not recommended. 1604 idle=halt: Halt is forced to be used for CPU idle. 1605 In such case C2/C3 won't be used again. 1606 idle=nomwait: Disable mwait for CPU C-states 1607 1608 ieee754= [MIPS] Select IEEE Std 754 conformance mode 1609 Format: { strict | legacy | 2008 | relaxed } 1610 Default: strict 1611 1612 Choose which programs will be accepted for execution 1613 based on the IEEE 754 NaN encoding(s) supported by 1614 the FPU and the NaN encoding requested with the value 1615 of an ELF file header flag individually set by each 1616 binary. Hardware implementations are permitted to 1617 support either or both of the legacy and the 2008 NaN 1618 encoding mode. 1619 1620 Available settings are as follows: 1621 strict accept binaries that request a NaN encoding 1622 supported by the FPU 1623 legacy only accept legacy-NaN binaries, if supported 1624 by the FPU 1625 2008 only accept 2008-NaN binaries, if supported 1626 by the FPU 1627 relaxed accept any binaries regardless of whether 1628 supported by the FPU 1629 1630 The FPU emulator is always able to support both NaN 1631 encodings, so if no FPU hardware is present or it has 1632 been disabled with 'nofpu', then the settings of 1633 'legacy' and '2008' strap the emulator accordingly, 1634 'relaxed' straps the emulator for both legacy-NaN and 1635 2008-NaN, whereas 'strict' enables legacy-NaN only on 1636 legacy processors and both NaN encodings on MIPS32 or 1637 MIPS64 CPUs. 1638 1639 The setting for ABS.fmt/NEG.fmt instruction execution 1640 mode generally follows that for the NaN encoding, 1641 except where unsupported by hardware. 1642 1643 ignore_loglevel [KNL] 1644 Ignore loglevel setting - this will print /all/ 1645 kernel messages to the console. Useful for debugging. 1646 We also add it as printk module parameter, so users 1647 could change it dynamically, usually by 1648 /sys/module/printk/parameters/ignore_loglevel. 1649 1650 ignore_rlimit_data 1651 Ignore RLIMIT_DATA setting for data mappings, 1652 print warning at first misuse. Can be changed via 1653 /sys/module/kernel/parameters/ignore_rlimit_data. 1654 1655 ihash_entries= [KNL] 1656 Set number of hash buckets for inode cache. 1657 1658 ima_appraise= [IMA] appraise integrity measurements 1659 Format: { "off" | "enforce" | "fix" | "log" } 1660 default: "enforce" 1661 1662 ima_appraise_tcb [IMA] Deprecated. Use ima_policy= instead. 1663 The builtin appraise policy appraises all files 1664 owned by uid=0. 1665 1666 ima_canonical_fmt [IMA] 1667 Use the canonical format for the binary runtime 1668 measurements, instead of host native format. 1669 1670 ima_hash= [IMA] 1671 Format: { md5 | sha1 | rmd160 | sha256 | sha384 1672 | sha512 | ... } 1673 default: "sha1" 1674 1675 The list of supported hash algorithms is defined 1676 in crypto/hash_info.h. 1677 1678 ima_policy= [IMA] 1679 The builtin policies to load during IMA setup. 1680 Format: "tcb | appraise_tcb | secure_boot | 1681 fail_securely" 1682 1683 The "tcb" policy measures all programs exec'd, files 1684 mmap'd for exec, and all files opened with the read 1685 mode bit set by either the effective uid (euid=0) or 1686 uid=0. 1687 1688 The "appraise_tcb" policy appraises the integrity of 1689 all files owned by root. 1690 1691 The "secure_boot" policy appraises the integrity 1692 of files (eg. kexec kernel image, kernel modules, 1693 firmware, policy, etc) based on file signatures. 1694 1695 The "fail_securely" policy forces file signature 1696 verification failure also on privileged mounted 1697 filesystems with the SB_I_UNVERIFIABLE_SIGNATURE 1698 flag. 1699 1700 ima_tcb [IMA] Deprecated. Use ima_policy= instead. 1701 Load a policy which meets the needs of the Trusted 1702 Computing Base. This means IMA will measure all 1703 programs exec'd, files mmap'd for exec, and all files 1704 opened for read by uid=0. 1705 1706 ima_template= [IMA] 1707 Select one of defined IMA measurements template formats. 1708 Formats: { "ima" | "ima-ng" | "ima-sig" } 1709 Default: "ima-ng" 1710 1711 ima_template_fmt= 1712 [IMA] Define a custom template format. 1713 Format: { "field1|...|fieldN" } 1714 1715 ima.ahash_minsize= [IMA] Minimum file size for asynchronous hash usage 1716 Format: <min_file_size> 1717 Set the minimal file size for using asynchronous hash. 1718 If left unspecified, ahash usage is disabled. 1719 1720 ahash performance varies for different data sizes on 1721 different crypto accelerators. This option can be used 1722 to achieve the best performance for a particular HW. 1723 1724 ima.ahash_bufsize= [IMA] Asynchronous hash buffer size 1725 Format: <bufsize> 1726 Set hashing buffer size. Default: 4k. 1727 1728 ahash performance varies for different chunk sizes on 1729 different crypto accelerators. This option can be used 1730 to achieve best performance for particular HW. 1731 1732 init= [KNL] 1733 Format: <full_path> 1734 Run specified binary instead of /sbin/init as init 1735 process. 1736 1737 initcall_debug [KNL] Trace initcalls as they are executed. Useful 1738 for working out where the kernel is dying during 1739 startup. 1740 1741 initcall_blacklist= [KNL] Do not execute a comma-separated list of 1742 initcall functions. Useful for debugging built-in 1743 modules and initcalls. 1744 1745 initrd= [BOOT] Specify the location of the initial ramdisk 1746 1747 init_on_alloc= [MM] Fill newly allocated pages and heap objects with 1748 zeroes. 1749 Format: 0 | 1 1750 Default set by CONFIG_INIT_ON_ALLOC_DEFAULT_ON. 1751 1752 init_on_free= [MM] Fill freed pages and heap objects with zeroes. 1753 Format: 0 | 1 1754 Default set by CONFIG_INIT_ON_FREE_DEFAULT_ON. 1755 1756 init_pkru= [x86] Specify the default memory protection keys rights 1757 register contents for all processes. 0x55555554 by 1758 default (disallow access to all but pkey 0). Can 1759 override in debugfs after boot. 1760 1761 inport.irq= [HW] Inport (ATI XL and Microsoft) busmouse driver 1762 Format: <irq> 1763 1764 int_pln_enable [x86] Enable power limit notification interrupt 1765 1766 integrity_audit=[IMA] 1767 Format: { "0" | "1" } 1768 0 -- basic integrity auditing messages. (Default) 1769 1 -- additional integrity auditing messages. 1770 1771 intel_iommu= [DMAR] Intel IOMMU driver (DMAR) option 1772 on 1773 Enable intel iommu driver. 1774 off 1775 Disable intel iommu driver. 1776 igfx_off [Default Off] 1777 By default, gfx is mapped as normal device. If a gfx 1778 device has a dedicated DMAR unit, the DMAR unit is 1779 bypassed by not enabling DMAR with this option. In 1780 this case, gfx device will use physical address for 1781 DMA. 1782 forcedac [x86_64] 1783 With this option iommu will not optimize to look 1784 for io virtual address below 32-bit forcing dual 1785 address cycle on pci bus for cards supporting greater 1786 than 32-bit addressing. The default is to look 1787 for translation below 32-bit and if not available 1788 then look in the higher range. 1789 strict [Default Off] 1790 With this option on every unmap_single operation will 1791 result in a hardware IOTLB flush operation as opposed 1792 to batching them for performance. 1793 sp_off [Default Off] 1794 By default, super page will be supported if Intel IOMMU 1795 has the capability. With this option, super page will 1796 not be supported. 1797 sm_on [Default Off] 1798 By default, scalable mode will be disabled even if the 1799 hardware advertises that it has support for the scalable 1800 mode translation. With this option set, scalable mode 1801 will be used on hardware which claims to support it. 1802 tboot_noforce [Default Off] 1803 Do not force the Intel IOMMU enabled under tboot. 1804 By default, tboot will force Intel IOMMU on, which 1805 could harm performance of some high-throughput 1806 devices like 40GBit network cards, even if identity 1807 mapping is enabled. 1808 Note that using this option lowers the security 1809 provided by tboot because it makes the system 1810 vulnerable to DMA attacks. 1811 nobounce [Default off] 1812 Disable bounce buffer for untrusted devices such as 1813 the Thunderbolt devices. This will treat the untrusted 1814 devices as the trusted ones, hence might expose security 1815 risks of DMA attacks. 1816 1817 intel_idle.max_cstate= [KNL,HW,ACPI,X86] 1818 0 disables intel_idle and fall back on acpi_idle. 1819 1 to 9 specify maximum depth of C-state. 1820 1821 intel_pstate= [X86] 1822 disable 1823 Do not enable intel_pstate as the default 1824 scaling driver for the supported processors 1825 passive 1826 Use intel_pstate as a scaling driver, but configure it 1827 to work with generic cpufreq governors (instead of 1828 enabling its internal governor). This mode cannot be 1829 used along with the hardware-managed P-states (HWP) 1830 feature. 1831 force 1832 Enable intel_pstate on systems that prohibit it by default 1833 in favor of acpi-cpufreq. Forcing the intel_pstate driver 1834 instead of acpi-cpufreq may disable platform features, such 1835 as thermal controls and power capping, that rely on ACPI 1836 P-States information being indicated to OSPM and therefore 1837 should be used with caution. This option does not work with 1838 processors that aren't supported by the intel_pstate driver 1839 or on platforms that use pcc-cpufreq instead of acpi-cpufreq. 1840 no_hwp 1841 Do not enable hardware P state control (HWP) 1842 if available. 1843 hwp_only 1844 Only load intel_pstate on systems which support 1845 hardware P state control (HWP) if available. 1846 support_acpi_ppc 1847 Enforce ACPI _PPC performance limits. If the Fixed ACPI 1848 Description Table, specifies preferred power management 1849 profile as "Enterprise Server" or "Performance Server", 1850 then this feature is turned on by default. 1851 per_cpu_perf_limits 1852 Allow per-logical-CPU P-State performance control limits using 1853 cpufreq sysfs interface 1854 1855 intremap= [X86-64, Intel-IOMMU] 1856 on enable Interrupt Remapping (default) 1857 off disable Interrupt Remapping 1858 nosid disable Source ID checking 1859 no_x2apic_optout 1860 BIOS x2APIC opt-out request will be ignored 1861 nopost disable Interrupt Posting 1862 1863 iomem= Disable strict checking of access to MMIO memory 1864 strict regions from userspace. 1865 relaxed 1866 1867 iommu= [x86] 1868 off 1869 force 1870 noforce 1871 biomerge 1872 panic 1873 nopanic 1874 merge 1875 nomerge 1876 soft 1877 pt [x86] 1878 nopt [x86] 1879 nobypass [PPC/POWERNV] 1880 Disable IOMMU bypass, using IOMMU for PCI devices. 1881 1882 iommu.strict= [ARM64] Configure TLB invalidation behaviour 1883 Format: { "0" | "1" } 1884 0 - Lazy mode. 1885 Request that DMA unmap operations use deferred 1886 invalidation of hardware TLBs, for increased 1887 throughput at the cost of reduced device isolation. 1888 Will fall back to strict mode if not supported by 1889 the relevant IOMMU driver. 1890 1 - Strict mode (default). 1891 DMA unmap operations invalidate IOMMU hardware TLBs 1892 synchronously. 1893 1894 iommu.passthrough= 1895 [ARM64, X86] Configure DMA to bypass the IOMMU by default. 1896 Format: { "0" | "1" } 1897 0 - Use IOMMU translation for DMA. 1898 1 - Bypass the IOMMU for DMA. 1899 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH. 1900 1901 io7= [HW] IO7 for Marvel based alpha systems 1902 See comment before marvel_specify_io7 in 1903 arch/alpha/kernel/core_marvel.c. 1904 1905 io_delay= [X86] I/O delay method 1906 0x80 1907 Standard port 0x80 based delay 1908 0xed 1909 Alternate port 0xed based delay (needed on some systems) 1910 udelay 1911 Simple two microseconds delay 1912 none 1913 No delay 1914 1915 ip= [IP_PNP] 1916 See Documentation/admin-guide/nfs/nfsroot.rst. 1917 1918 ipcmni_extend [KNL] Extend the maximum number of unique System V 1919 IPC identifiers from 32,768 to 16,777,216. 1920 1921 irqaffinity= [SMP] Set the default irq affinity mask 1922 The argument is a cpu list, as described above. 1923 1924 irqchip.gicv2_force_probe= 1925 [ARM, ARM64] 1926 Format: <bool> 1927 Force the kernel to look for the second 4kB page 1928 of a GICv2 controller even if the memory range 1929 exposed by the device tree is too small. 1930 1931 irqchip.gicv3_nolpi= 1932 [ARM, ARM64] 1933 Force the kernel to ignore the availability of 1934 LPIs (and by consequence ITSs). Intended for system 1935 that use the kernel as a bootloader, and thus want 1936 to let secondary kernels in charge of setting up 1937 LPIs. 1938 1939 irqchip.gicv3_pseudo_nmi= [ARM64] 1940 Enables support for pseudo-NMIs in the kernel. This 1941 requires the kernel to be built with 1942 CONFIG_ARM64_PSEUDO_NMI. 1943 1944 irqfixup [HW] 1945 When an interrupt is not handled search all handlers 1946 for it. Intended to get systems with badly broken 1947 firmware running. 1948 1949 irqpoll [HW] 1950 When an interrupt is not handled search all handlers 1951 for it. Also check all handlers each timer 1952 interrupt. Intended to get systems with badly broken 1953 firmware running. 1954 1955 isapnp= [ISAPNP] 1956 Format: <RDP>,<reset>,<pci_scan>,<verbosity> 1957 1958 isolcpus= [KNL,SMP,ISOL] Isolate a given set of CPUs from disturbance. 1959 [Deprecated - use cpusets instead] 1960 Format: [flag-list,]<cpu-list> 1961 1962 Specify one or more CPUs to isolate from disturbances 1963 specified in the flag list (default: domain): 1964 1965 nohz 1966 Disable the tick when a single task runs. 1967 1968 A residual 1Hz tick is offloaded to workqueues, which you 1969 need to affine to housekeeping through the global 1970 workqueue's affinity configured via the 1971 /sys/devices/virtual/workqueue/cpumask sysfs file, or 1972 by using the 'domain' flag described below. 1973 1974 NOTE: by default the global workqueue runs on all CPUs, 1975 so to protect individual CPUs the 'cpumask' file has to 1976 be configured manually after bootup. 1977 1978 domain 1979 Isolate from the general SMP balancing and scheduling 1980 algorithms. Note that performing domain isolation this way 1981 is irreversible: it's not possible to bring back a CPU to 1982 the domains once isolated through isolcpus. It's strongly 1983 advised to use cpusets instead to disable scheduler load 1984 balancing through the "cpuset.sched_load_balance" file. 1985 It offers a much more flexible interface where CPUs can 1986 move in and out of an isolated set anytime. 1987 1988 You can move a process onto or off an "isolated" CPU via 1989 the CPU affinity syscalls or cpuset. 1990 <cpu number> begins at 0 and the maximum value is 1991 "number of CPUs in system - 1". 1992 1993 managed_irq 1994 1995 Isolate from being targeted by managed interrupts 1996 which have an interrupt mask containing isolated 1997 CPUs. The affinity of managed interrupts is 1998 handled by the kernel and cannot be changed via 1999 the /proc/irq/* interfaces. 2000 2001 This isolation is best effort and only effective 2002 if the automatically assigned interrupt mask of a 2003 device queue contains isolated and housekeeping 2004 CPUs. If housekeeping CPUs are online then such 2005 interrupts are directed to the housekeeping CPU 2006 so that IO submitted on the housekeeping CPU 2007 cannot disturb the isolated CPU. 2008 2009 If a queue's affinity mask contains only isolated 2010 CPUs then this parameter has no effect on the 2011 interrupt routing decision, though interrupts are 2012 only delivered when tasks running on those 2013 isolated CPUs submit IO. IO submitted on 2014 housekeeping CPUs has no influence on those 2015 queues. 2016 2017 The format of <cpu-list> is described above. 2018 2019 iucv= [HW,NET] 2020 2021 ivrs_ioapic [HW,X86_64] 2022 Provide an override to the IOAPIC-ID<->DEVICE-ID 2023 mapping provided in the IVRS ACPI table. For 2024 example, to map IOAPIC-ID decimal 10 to 2025 PCI device 00:14.0 write the parameter as: 2026 ivrs_ioapic[10]=00:14.0 2027 2028 ivrs_hpet [HW,X86_64] 2029 Provide an override to the HPET-ID<->DEVICE-ID 2030 mapping provided in the IVRS ACPI table. For 2031 example, to map HPET-ID decimal 0 to 2032 PCI device 00:14.0 write the parameter as: 2033 ivrs_hpet[0]=00:14.0 2034 2035 ivrs_acpihid [HW,X86_64] 2036 Provide an override to the ACPI-HID:UID<->DEVICE-ID 2037 mapping provided in the IVRS ACPI table. For 2038 example, to map UART-HID:UID AMD0020:0 to 2039 PCI device 00:14.5 write the parameter as: 2040 ivrs_acpihid[00:14.5]=AMD0020:0 2041 2042 js= [HW,JOY] Analog joystick 2043 See Documentation/input/joydev/joystick.rst. 2044 2045 nokaslr [KNL] 2046 When CONFIG_RANDOMIZE_BASE is set, this disables 2047 kernel and module base offset ASLR (Address Space 2048 Layout Randomization). 2049 2050 kasan_multi_shot 2051 [KNL] Enforce KASAN (Kernel Address Sanitizer) to print 2052 report on every invalid memory access. Without this 2053 parameter KASAN will print report only for the first 2054 invalid access. 2055 2056 keepinitrd [HW,ARM] 2057 2058 kernelcore= [KNL,X86,IA-64,PPC] 2059 Format: nn[KMGTPE] | nn% | "mirror" 2060 This parameter specifies the amount of memory usable by 2061 the kernel for non-movable allocations. The requested 2062 amount is spread evenly throughout all nodes in the 2063 system as ZONE_NORMAL. The remaining memory is used for 2064 movable memory in its own zone, ZONE_MOVABLE. In the 2065 event, a node is too small to have both ZONE_NORMAL and 2066 ZONE_MOVABLE, kernelcore memory will take priority and 2067 other nodes will have a larger ZONE_MOVABLE. 2068 2069 ZONE_MOVABLE is used for the allocation of pages that 2070 may be reclaimed or moved by the page migration 2071 subsystem. Note that allocations like PTEs-from-HighMem 2072 still use the HighMem zone if it exists, and the Normal 2073 zone if it does not. 2074 2075 It is possible to specify the exact amount of memory in 2076 the form of "nn[KMGTPE]", a percentage of total system 2077 memory in the form of "nn%", or "mirror". If "mirror" 2078 option is specified, mirrored (reliable) memory is used 2079 for non-movable allocations and remaining memory is used 2080 for Movable pages. "nn[KMGTPE]", "nn%", and "mirror" 2081 are exclusive, so you cannot specify multiple forms. 2082 2083 kgdbdbgp= [KGDB,HW] kgdb over EHCI usb debug port. 2084 Format: <Controller#>[,poll interval] 2085 The controller # is the number of the ehci usb debug 2086 port as it is probed via PCI. The poll interval is 2087 optional and is the number seconds in between 2088 each poll cycle to the debug port in case you need 2089 the functionality for interrupting the kernel with 2090 gdb or control-c on the dbgp connection. When 2091 not using this parameter you use sysrq-g to break into 2092 the kernel debugger. 2093 2094 kgdboc= [KGDB,HW] kgdb over consoles. 2095 Requires a tty driver that supports console polling, 2096 or a supported polling keyboard driver (non-usb). 2097 Serial only format: <serial_device>[,baud] 2098 keyboard only format: kbd 2099 keyboard and serial format: kbd,<serial_device>[,baud] 2100 Optional Kernel mode setting: 2101 kms, kbd format: kms,kbd 2102 kms, kbd and serial format: kms,kbd,<ser_dev>[,baud] 2103 2104 kgdbwait [KGDB] Stop kernel execution and enter the 2105 kernel debugger at the earliest opportunity. 2106 2107 kmac= [MIPS] korina ethernet MAC address. 2108 Configure the RouterBoard 532 series on-chip 2109 Ethernet adapter MAC address. 2110 2111 kmemleak= [KNL] Boot-time kmemleak enable/disable 2112 Valid arguments: on, off 2113 Default: on 2114 Built with CONFIG_DEBUG_KMEMLEAK_DEFAULT_OFF=y, 2115 the default is off. 2116 2117 kprobe_event=[probe-list] 2118 [FTRACE] Add kprobe events and enable at boot time. 2119 The probe-list is a semicolon delimited list of probe 2120 definitions. Each definition is same as kprobe_events 2121 interface, but the parameters are comma delimited. 2122 For example, to add a kprobe event on vfs_read with 2123 arg1 and arg2, add to the command line; 2124 2125 kprobe_event=p,vfs_read,$arg1,$arg2 2126 2127 See also Documentation/trace/kprobetrace.rst "Kernel 2128 Boot Parameter" section. 2129 2130 kpti= [ARM64] Control page table isolation of user 2131 and kernel address spaces. 2132 Default: enabled on cores which need mitigation. 2133 0: force disabled 2134 1: force enabled 2135 2136 kvm.ignore_msrs=[KVM] Ignore guest accesses to unhandled MSRs. 2137 Default is 0 (don't ignore, but inject #GP) 2138 2139 kvm.enable_vmware_backdoor=[KVM] Support VMware backdoor PV interface. 2140 Default is false (don't support). 2141 2142 kvm.mmu_audit= [KVM] This is a R/W parameter which allows audit 2143 KVM MMU at runtime. 2144 Default is 0 (off) 2145 2146 kvm.nx_huge_pages= 2147 [KVM] Controls the software workaround for the 2148 X86_BUG_ITLB_MULTIHIT bug. 2149 force : Always deploy workaround. 2150 off : Never deploy workaround. 2151 auto : Deploy workaround based on the presence of 2152 X86_BUG_ITLB_MULTIHIT. 2153 2154 Default is 'auto'. 2155 2156 If the software workaround is enabled for the host, 2157 guests do need not to enable it for nested guests. 2158 2159 kvm.nx_huge_pages_recovery_ratio= 2160 [KVM] Controls how many 4KiB pages are periodically zapped 2161 back to huge pages. 0 disables the recovery, otherwise if 2162 the value is N KVM will zap 1/Nth of the 4KiB pages every 2163 minute. The default is 60. 2164 2165 kvm-amd.nested= [KVM,AMD] Allow nested virtualization in KVM/SVM. 2166 Default is 1 (enabled) 2167 2168 kvm-amd.npt= [KVM,AMD] Disable nested paging (virtualized MMU) 2169 for all guests. 2170 Default is 1 (enabled) if in 64-bit or 32-bit PAE mode. 2171 2172 kvm-arm.vgic_v3_group0_trap= 2173 [KVM,ARM] Trap guest accesses to GICv3 group-0 2174 system registers 2175 2176 kvm-arm.vgic_v3_group1_trap= 2177 [KVM,ARM] Trap guest accesses to GICv3 group-1 2178 system registers 2179 2180 kvm-arm.vgic_v3_common_trap= 2181 [KVM,ARM] Trap guest accesses to GICv3 common 2182 system registers 2183 2184 kvm-arm.vgic_v4_enable= 2185 [KVM,ARM] Allow use of GICv4 for direct injection of 2186 LPIs. 2187 2188 kvm-intel.ept= [KVM,Intel] Disable extended page tables 2189 (virtualized MMU) support on capable Intel chips. 2190 Default is 1 (enabled) 2191 2192 kvm-intel.emulate_invalid_guest_state= 2193 [KVM,Intel] Enable emulation of invalid guest states 2194 Default is 0 (disabled) 2195 2196 kvm-intel.flexpriority= 2197 [KVM,Intel] Disable FlexPriority feature (TPR shadow). 2198 Default is 1 (enabled) 2199 2200 kvm-intel.nested= 2201 [KVM,Intel] Enable VMX nesting (nVMX). 2202 Default is 0 (disabled) 2203 2204 kvm-intel.unrestricted_guest= 2205 [KVM,Intel] Disable unrestricted guest feature 2206 (virtualized real and unpaged mode) on capable 2207 Intel chips. Default is 1 (enabled) 2208 2209 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault 2210 CVE-2018-3620. 2211 2212 Valid arguments: never, cond, always 2213 2214 always: L1D cache flush on every VMENTER. 2215 cond: Flush L1D on VMENTER only when the code between 2216 VMEXIT and VMENTER can leak host memory. 2217 never: Disables the mitigation 2218 2219 Default is cond (do L1 cache flush in specific instances) 2220 2221 kvm-intel.vpid= [KVM,Intel] Disable Virtual Processor Identification 2222 feature (tagged TLBs) on capable Intel chips. 2223 Default is 1 (enabled) 2224 2225 l1tf= [X86] Control mitigation of the L1TF vulnerability on 2226 affected CPUs 2227 2228 The kernel PTE inversion protection is unconditionally 2229 enabled and cannot be disabled. 2230 2231 full 2232 Provides all available mitigations for the 2233 L1TF vulnerability. Disables SMT and 2234 enables all mitigations in the 2235 hypervisors, i.e. unconditional L1D flush. 2236 2237 SMT control and L1D flush control via the 2238 sysfs interface is still possible after 2239 boot. Hypervisors will issue a warning 2240 when the first VM is started in a 2241 potentially insecure configuration, 2242 i.e. SMT enabled or L1D flush disabled. 2243 2244 full,force 2245 Same as 'full', but disables SMT and L1D 2246 flush runtime control. Implies the 2247 'nosmt=force' command line option. 2248 (i.e. sysfs control of SMT is disabled.) 2249 2250 flush 2251 Leaves SMT enabled and enables the default 2252 hypervisor mitigation, i.e. conditional 2253 L1D flush. 2254 2255 SMT control and L1D flush control via the 2256 sysfs interface is still possible after 2257 boot. Hypervisors will issue a warning 2258 when the first VM is started in a 2259 potentially insecure configuration, 2260 i.e. SMT enabled or L1D flush disabled. 2261 2262 flush,nosmt 2263 2264 Disables SMT and enables the default 2265 hypervisor mitigation. 2266 2267 SMT control and L1D flush control via the 2268 sysfs interface is still possible after 2269 boot. Hypervisors will issue a warning 2270 when the first VM is started in a 2271 potentially insecure configuration, 2272 i.e. SMT enabled or L1D flush disabled. 2273 2274 flush,nowarn 2275 Same as 'flush', but hypervisors will not 2276 warn when a VM is started in a potentially 2277 insecure configuration. 2278 2279 off 2280 Disables hypervisor mitigations and doesn't 2281 emit any warnings. 2282 It also drops the swap size and available 2283 RAM limit restriction on both hypervisor and 2284 bare metal. 2285 2286 Default is 'flush'. 2287 2288 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst 2289 2290 l2cr= [PPC] 2291 2292 l3cr= [PPC] 2293 2294 lapic [X86-32,APIC] Enable the local APIC even if BIOS 2295 disabled it. 2296 2297 lapic= [x86,APIC] "notscdeadline" Do not use TSC deadline 2298 value for LAPIC timer one-shot implementation. Default 2299 back to the programmable timer unit in the LAPIC. 2300 2301 lapic_timer_c2_ok [X86,APIC] trust the local apic timer 2302 in C2 power state. 2303 2304 libata.dma= [LIBATA] DMA control 2305 libata.dma=0 Disable all PATA and SATA DMA 2306 libata.dma=1 PATA and SATA Disk DMA only 2307 libata.dma=2 ATAPI (CDROM) DMA only 2308 libata.dma=4 Compact Flash DMA only 2309 Combinations also work, so libata.dma=3 enables DMA 2310 for disks and CDROMs, but not CFs. 2311 2312 libata.ignore_hpa= [LIBATA] Ignore HPA limit 2313 libata.ignore_hpa=0 keep BIOS limits (default) 2314 libata.ignore_hpa=1 ignore limits, using full disk 2315 2316 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume 2317 when set. 2318 Format: <int> 2319 2320 libata.force= [LIBATA] Force configurations. The format is comma 2321 separated list of "[ID:]VAL" where ID is 2322 PORT[.DEVICE]. PORT and DEVICE are decimal numbers 2323 matching port, link or device. Basically, it matches 2324 the ATA ID string printed on console by libata. If 2325 the whole ID part is omitted, the last PORT and DEVICE 2326 values are used. If ID hasn't been specified yet, the 2327 configuration applies to all ports, links and devices. 2328 2329 If only DEVICE is omitted, the parameter applies to 2330 the port and all links and devices behind it. DEVICE 2331 number of 0 either selects the first device or the 2332 first fan-out link behind PMP device. It does not 2333 select the host link. DEVICE number of 15 selects the 2334 host link and device attached to it. 2335 2336 The VAL specifies the configuration to force. As long 2337 as there's no ambiguity shortcut notation is allowed. 2338 For example, both 1.5 and 1.5G would work for 1.5Gbps. 2339 The following configurations can be forced. 2340 2341 * Cable type: 40c, 80c, short40c, unk, ign or sata. 2342 Any ID with matching PORT is used. 2343 2344 * SATA link speed limit: 1.5Gbps or 3.0Gbps. 2345 2346 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7]. 2347 udma[/][16,25,33,44,66,100,133] notation is also 2348 allowed. 2349 2350 * [no]ncq: Turn on or off NCQ. 2351 2352 * [no]ncqtrim: Turn off queued DSM TRIM. 2353 2354 * nohrst, nosrst, norst: suppress hard, soft 2355 and both resets. 2356 2357 * rstonce: only attempt one reset during 2358 hot-unplug link recovery 2359 2360 * dump_id: dump IDENTIFY data. 2361 2362 * atapi_dmadir: Enable ATAPI DMADIR bridge support 2363 2364 * disable: Disable this device. 2365 2366 If there are multiple matching configurations changing 2367 the same attribute, the last one is used. 2368 2369 memblock=debug [KNL] Enable memblock debug messages. 2370 2371 load_ramdisk= [RAM] List of ramdisks to load from floppy 2372 See Documentation/admin-guide/blockdev/ramdisk.rst. 2373 2374 lockd.nlm_grace_period=P [NFS] Assign grace period. 2375 Format: <integer> 2376 2377 lockd.nlm_tcpport=N [NFS] Assign TCP port. 2378 Format: <integer> 2379 2380 lockd.nlm_timeout=T [NFS] Assign timeout value. 2381 Format: <integer> 2382 2383 lockd.nlm_udpport=M [NFS] Assign UDP port. 2384 Format: <integer> 2385 2386 lockdown= [SECURITY] 2387 { integrity | confidentiality } 2388 Enable the kernel lockdown feature. If set to 2389 integrity, kernel features that allow userland to 2390 modify the running kernel are disabled. If set to 2391 confidentiality, kernel features that allow userland 2392 to extract confidential information from the kernel 2393 are also disabled. 2394 2395 locktorture.nreaders_stress= [KNL] 2396 Set the number of locking read-acquisition kthreads. 2397 Defaults to being automatically set based on the 2398 number of online CPUs. 2399 2400 locktorture.nwriters_stress= [KNL] 2401 Set the number of locking write-acquisition kthreads. 2402 2403 locktorture.onoff_holdoff= [KNL] 2404 Set time (s) after boot for CPU-hotplug testing. 2405 2406 locktorture.onoff_interval= [KNL] 2407 Set time (s) between CPU-hotplug operations, or 2408 zero to disable CPU-hotplug testing. 2409 2410 locktorture.shuffle_interval= [KNL] 2411 Set task-shuffle interval (jiffies). Shuffling 2412 tasks allows some CPUs to go into dyntick-idle 2413 mode during the locktorture test. 2414 2415 locktorture.shutdown_secs= [KNL] 2416 Set time (s) after boot system shutdown. This 2417 is useful for hands-off automated testing. 2418 2419 locktorture.stat_interval= [KNL] 2420 Time (s) between statistics printk()s. 2421 2422 locktorture.stutter= [KNL] 2423 Time (s) to stutter testing, for example, 2424 specifying five seconds causes the test to run for 2425 five seconds, wait for five seconds, and so on. 2426 This tests the locking primitive's ability to 2427 transition abruptly to and from idle. 2428 2429 locktorture.torture_type= [KNL] 2430 Specify the locking implementation to test. 2431 2432 locktorture.verbose= [KNL] 2433 Enable additional printk() statements. 2434 2435 logibm.irq= [HW,MOUSE] Logitech Bus Mouse Driver 2436 Format: <irq> 2437 2438 loglevel= All Kernel Messages with a loglevel smaller than the 2439 console loglevel will be printed to the console. It can 2440 also be changed with klogd or other programs. The 2441 loglevels are defined as follows: 2442 2443 0 (KERN_EMERG) system is unusable 2444 1 (KERN_ALERT) action must be taken immediately 2445 2 (KERN_CRIT) critical conditions 2446 3 (KERN_ERR) error conditions 2447 4 (KERN_WARNING) warning conditions 2448 5 (KERN_NOTICE) normal but significant condition 2449 6 (KERN_INFO) informational 2450 7 (KERN_DEBUG) debug-level messages 2451 2452 log_buf_len=n[KMG] Sets the size of the printk ring buffer, 2453 in bytes. n must be a power of two and greater 2454 than the minimal size. The minimal size is defined 2455 by LOG_BUF_SHIFT kernel config parameter. There is 2456 also CONFIG_LOG_CPU_MAX_BUF_SHIFT config parameter 2457 that allows to increase the default size depending on 2458 the number of CPUs. See init/Kconfig for more details. 2459 2460 logo.nologo [FB] Disables display of the built-in Linux logo. 2461 This may be used to provide more screen space for 2462 kernel log messages and is useful when debugging 2463 kernel boot problems. 2464 2465 lp=0 [LP] Specify parallel ports to use, e.g, 2466 lp=port[,port...] lp=none,parport0 (lp0 not configured, lp1 uses 2467 lp=reset first parallel port). 'lp=0' disables the 2468 lp=auto printer driver. 'lp=reset' (which can be 2469 specified in addition to the ports) causes 2470 attached printers to be reset. Using 2471 lp=port1,port2,... specifies the parallel ports 2472 to associate lp devices with, starting with 2473 lp0. A port specification may be 'none' to skip 2474 that lp device, or a parport name such as 2475 'parport0'. Specifying 'lp=auto' instead of a 2476 port specification list means that device IDs 2477 from each port should be examined, to see if 2478 an IEEE 1284-compliant printer is attached; if 2479 so, the driver will manage that printer. 2480 See also header of drivers/char/lp.c. 2481 2482 lpj=n [KNL] 2483 Sets loops_per_jiffy to given constant, thus avoiding 2484 time-consuming boot-time autodetection (up to 250 ms per 2485 CPU). 0 enables autodetection (default). To determine 2486 the correct value for your kernel, boot with normal 2487 autodetection and see what value is printed. Note that 2488 on SMP systems the preset will be applied to all CPUs, 2489 which is likely to cause problems if your CPUs need 2490 significantly divergent settings. An incorrect value 2491 will cause delays in the kernel to be wrong, leading to 2492 unpredictable I/O errors and other breakage. Although 2493 unlikely, in the extreme case this might damage your 2494 hardware. 2495 2496 ltpc= [NET] 2497 Format: <io>,<irq>,<dma> 2498 2499 lsm.debug [SECURITY] Enable LSM initialization debugging output. 2500 2501 lsm=lsm1,...,lsmN 2502 [SECURITY] Choose order of LSM initialization. This 2503 overrides CONFIG_LSM, and the "security=" parameter. 2504 2505 machvec= [IA-64] Force the use of a particular machine-vector 2506 (machvec) in a generic kernel. 2507 Example: machvec=hpzx1 2508 2509 machtype= [Loongson] Share the same kernel image file between different 2510 yeeloong laptop. 2511 Example: machtype=lemote-yeeloong-2f-7inch 2512 2513 max_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory greater 2514 than or equal to this physical address is ignored. 2515 2516 maxcpus= [SMP] Maximum number of processors that an SMP kernel 2517 will bring up during bootup. maxcpus=n : n >= 0 limits 2518 the kernel to bring up 'n' processors. Surely after 2519 bootup you can bring up the other plugged cpu by executing 2520 "echo 1 > /sys/devices/system/cpu/cpuX/online". So maxcpus 2521 only takes effect during system bootup. 2522 While n=0 is a special case, it is equivalent to "nosmp", 2523 which also disables the IO APIC. 2524 2525 max_loop= [LOOP] The number of loop block devices that get 2526 (loop.max_loop) unconditionally pre-created at init time. The default 2527 number is configured by BLK_DEV_LOOP_MIN_COUNT. Instead 2528 of statically allocating a predefined number, loop 2529 devices can be requested on-demand with the 2530 /dev/loop-control interface. 2531 2532 mce [X86-32] Machine Check Exception 2533 2534 mce=option [X86-64] See Documentation/x86/x86_64/boot-options.rst 2535 2536 md= [HW] RAID subsystems devices and level 2537 See Documentation/admin-guide/md.rst. 2538 2539 mdacon= [MDA] 2540 Format: <first>,<last> 2541 Specifies range of consoles to be captured by the MDA. 2542 2543 mds= [X86,INTEL] 2544 Control mitigation for the Micro-architectural Data 2545 Sampling (MDS) vulnerability. 2546 2547 Certain CPUs are vulnerable to an exploit against CPU 2548 internal buffers which can forward information to a 2549 disclosure gadget under certain conditions. 2550 2551 In vulnerable processors, the speculatively 2552 forwarded data can be used in a cache side channel 2553 attack, to access data to which the attacker does 2554 not have direct access. 2555 2556 This parameter controls the MDS mitigation. The 2557 options are: 2558 2559 full - Enable MDS mitigation on vulnerable CPUs 2560 full,nosmt - Enable MDS mitigation and disable 2561 SMT on vulnerable CPUs 2562 off - Unconditionally disable MDS mitigation 2563 2564 On TAA-affected machines, mds=off can be prevented by 2565 an active TAA mitigation as both vulnerabilities are 2566 mitigated with the same mechanism so in order to disable 2567 this mitigation, you need to specify tsx_async_abort=off 2568 too. 2569 2570 Not specifying this option is equivalent to 2571 mds=full. 2572 2573 For details see: Documentation/admin-guide/hw-vuln/mds.rst 2574 2575 mem=nn[KMG] [KNL,BOOT] Force usage of a specific amount of memory 2576 Amount of memory to be used in cases as follows: 2577 2578 1 for test; 2579 2 when the kernel is not able to see the whole system memory; 2580 3 memory that lies after 'mem=' boundary is excluded from 2581 the hypervisor, then assigned to KVM guests. 2582 2583 [X86] Work as limiting max address. Use together 2584 with memmap= to avoid physical address space collisions. 2585 Without memmap= PCI devices could be placed at addresses 2586 belonging to unused RAM. 2587 2588 Note that this only takes effects during boot time since 2589 in above case 3, memory may need be hot added after boot 2590 if system memory of hypervisor is not sufficient. 2591 2592 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel 2593 memory. 2594 2595 memchunk=nn[KMG] 2596 [KNL,SH] Allow user to override the default size for 2597 per-device physically contiguous DMA buffers. 2598 2599 memhp_default_state=online/offline 2600 [KNL] Set the initial state for the memory hotplug 2601 onlining policy. If not specified, the default value is 2602 set according to the 2603 CONFIG_MEMORY_HOTPLUG_DEFAULT_ONLINE kernel config 2604 option. 2605 See Documentation/admin-guide/mm/memory-hotplug.rst. 2606 2607 memmap=exactmap [KNL,X86] Enable setting of an exact 2608 E820 memory map, as specified by the user. 2609 Such memmap=exactmap lines can be constructed based on 2610 BIOS output or other requirements. See the memmap=nn@ss 2611 option description. 2612 2613 memmap=nn[KMG]@ss[KMG] 2614 [KNL] Force usage of a specific region of memory. 2615 Region of memory to be used is from ss to ss+nn. 2616 If @ss[KMG] is omitted, it is equivalent to mem=nn[KMG], 2617 which limits max address to nn[KMG]. 2618 Multiple different regions can be specified, 2619 comma delimited. 2620 Example: 2621 memmap=100M@2G,100M#3G,1G!1024G 2622 2623 memmap=nn[KMG]#ss[KMG] 2624 [KNL,ACPI] Mark specific memory as ACPI data. 2625 Region of memory to be marked is from ss to ss+nn. 2626 2627 memmap=nn[KMG]$ss[KMG] 2628 [KNL,ACPI] Mark specific memory as reserved. 2629 Region of memory to be reserved is from ss to ss+nn. 2630 Example: Exclude memory from 0x18690000-0x1869ffff 2631 memmap=64K$0x18690000 2632 or 2633 memmap=0x10000$0x18690000 2634 Some bootloaders may need an escape character before '$', 2635 like Grub2, otherwise '$' and the following number 2636 will be eaten. 2637 2638 memmap=nn[KMG]!ss[KMG] 2639 [KNL,X86] Mark specific memory as protected. 2640 Region of memory to be used, from ss to ss+nn. 2641 The memory region may be marked as e820 type 12 (0xc) 2642 and is NVDIMM or ADR memory. 2643 2644 memmap=<size>%<offset>-<oldtype>+<newtype> 2645 [KNL,ACPI] Convert memory within the specified region 2646 from <oldtype> to <newtype>. If "-<oldtype>" is left 2647 out, the whole region will be marked as <newtype>, 2648 even if previously unavailable. If "+<newtype>" is left 2649 out, matching memory will be removed. Types are 2650 specified as e820 types, e.g., 1 = RAM, 2 = reserved, 2651 3 = ACPI, 12 = PRAM. 2652 2653 memory_corruption_check=0/1 [X86] 2654 Some BIOSes seem to corrupt the first 64k of 2655 memory when doing things like suspend/resume. 2656 Setting this option will scan the memory 2657 looking for corruption. Enabling this will 2658 both detect corruption and prevent the kernel 2659 from using the memory being corrupted. 2660 However, its intended as a diagnostic tool; if 2661 repeatable BIOS-originated corruption always 2662 affects the same memory, you can use memmap= 2663 to prevent the kernel from using that memory. 2664 2665 memory_corruption_check_size=size [X86] 2666 By default it checks for corruption in the low 2667 64k, making this memory unavailable for normal 2668 use. Use this parameter to scan for 2669 corruption in more or less memory. 2670 2671 memory_corruption_check_period=seconds [X86] 2672 By default it checks for corruption every 60 2673 seconds. Use this parameter to check at some 2674 other rate. 0 disables periodic checking. 2675 2676 memtest= [KNL,X86,ARM,PPC] Enable memtest 2677 Format: <integer> 2678 default : 0 <disable> 2679 Specifies the number of memtest passes to be 2680 performed. Each pass selects another test 2681 pattern from a given set of patterns. Memtest 2682 fills the memory with this pattern, validates 2683 memory contents and reserves bad memory 2684 regions that are detected. 2685 2686 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control 2687 Valid arguments: on, off 2688 Default (depends on kernel configuration option): 2689 on (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=y) 2690 off (CONFIG_AMD_MEM_ENCRYPT_ACTIVE_BY_DEFAULT=n) 2691 mem_encrypt=on: Activate SME 2692 mem_encrypt=off: Do not activate SME 2693 2694 Refer to Documentation/virt/kvm/amd-memory-encryption.rst 2695 for details on when memory encryption can be activated. 2696 2697 mem_sleep_default= [SUSPEND] Default system suspend mode: 2698 s2idle - Suspend-To-Idle 2699 shallow - Power-On Suspend or equivalent (if supported) 2700 deep - Suspend-To-RAM or equivalent (if supported) 2701 See Documentation/admin-guide/pm/sleep-states.rst. 2702 2703 meye.*= [HW] Set MotionEye Camera parameters 2704 See Documentation/media/v4l-drivers/meye.rst. 2705 2706 mfgpt_irq= [IA-32] Specify the IRQ to use for the 2707 Multi-Function General Purpose Timers on AMD Geode 2708 platforms. 2709 2710 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when 2711 the BIOS has incorrectly applied a workaround. TinyBIOS 2712 version 0.98 is known to be affected, 0.99 fixes the 2713 problem by letting the user disable the workaround. 2714 2715 mga= [HW,DRM] 2716 2717 min_addr=nn[KMG] [KNL,BOOT,ia64] All physical memory below this 2718 physical address is ignored. 2719 2720 mini2440= [ARM,HW,KNL] 2721 Format:[0..2][b][c][t] 2722 Default: "0tb" 2723 MINI2440 configuration specification: 2724 0 - The attached screen is the 3.5" TFT 2725 1 - The attached screen is the 7" TFT 2726 2 - The VGA Shield is attached (1024x768) 2727 Leaving out the screen size parameter will not load 2728 the TFT driver, and the framebuffer will be left 2729 unconfigured. 2730 b - Enable backlight. The TFT backlight pin will be 2731 linked to the kernel VESA blanking code and a GPIO 2732 LED. This parameter is not necessary when using the 2733 VGA shield. 2734 c - Enable the s3c camera interface. 2735 t - Reserved for enabling touchscreen support. The 2736 touchscreen support is not enabled in the mainstream 2737 kernel as of 2.6.30, a preliminary port can be found 2738 in the "bleeding edge" mini2440 support kernel at 2739 http://repo.or.cz/w/linux-2.6/mini2440.git 2740 2741 mitigations= 2742 [X86,PPC,S390,ARM64] Control optional mitigations for 2743 CPU vulnerabilities. This is a set of curated, 2744 arch-independent options, each of which is an 2745 aggregation of existing arch-specific options. 2746 2747 off 2748 Disable all optional CPU mitigations. This 2749 improves system performance, but it may also 2750 expose users to several CPU vulnerabilities. 2751 Equivalent to: nopti [X86,PPC] 2752 kpti=0 [ARM64] 2753 nospectre_v1 [X86,PPC] 2754 nobp=0 [S390] 2755 nospectre_v2 [X86,PPC,S390,ARM64] 2756 spectre_v2_user=off [X86] 2757 spec_store_bypass_disable=off [X86,PPC] 2758 ssbd=force-off [ARM64] 2759 l1tf=off [X86] 2760 mds=off [X86] 2761 tsx_async_abort=off [X86] 2762 kvm.nx_huge_pages=off [X86] 2763 2764 Exceptions: 2765 This does not have any effect on 2766 kvm.nx_huge_pages when 2767 kvm.nx_huge_pages=force. 2768 2769 auto (default) 2770 Mitigate all CPU vulnerabilities, but leave SMT 2771 enabled, even if it's vulnerable. This is for 2772 users who don't want to be surprised by SMT 2773 getting disabled across kernel upgrades, or who 2774 have other ways of avoiding SMT-based attacks. 2775 Equivalent to: (default behavior) 2776 2777 auto,nosmt 2778 Mitigate all CPU vulnerabilities, disabling SMT 2779 if needed. This is for users who always want to 2780 be fully mitigated, even if it means losing SMT. 2781 Equivalent to: l1tf=flush,nosmt [X86] 2782 mds=full,nosmt [X86] 2783 tsx_async_abort=full,nosmt [X86] 2784 2785 mminit_loglevel= 2786 [KNL] When CONFIG_DEBUG_MEMORY_INIT is set, this 2787 parameter allows control of the logging verbosity for 2788 the additional memory initialisation checks. A value 2789 of 0 disables mminit logging and a level of 4 will 2790 log everything. Information is printed at KERN_DEBUG 2791 so loglevel=8 may also need to be specified. 2792 2793 module.sig_enforce 2794 [KNL] When CONFIG_MODULE_SIG is set, this means that 2795 modules without (valid) signatures will fail to load. 2796 Note that if CONFIG_MODULE_SIG_FORCE is set, that 2797 is always true, so this option does nothing. 2798 2799 module_blacklist= [KNL] Do not load a comma-separated list of 2800 modules. Useful for debugging problem modules. 2801 2802 mousedev.tap_time= 2803 [MOUSE] Maximum time between finger touching and 2804 leaving touchpad surface for touch to be considered 2805 a tap and be reported as a left button click (for 2806 touchpads working in absolute mode only). 2807 Format: <msecs> 2808 mousedev.xres= [MOUSE] Horizontal screen resolution, used for devices 2809 reporting absolute coordinates, such as tablets 2810 mousedev.yres= [MOUSE] Vertical screen resolution, used for devices 2811 reporting absolute coordinates, such as tablets 2812 2813 movablecore= [KNL,X86,IA-64,PPC] 2814 Format: nn[KMGTPE] | nn% 2815 This parameter is the complement to kernelcore=, it 2816 specifies the amount of memory used for migratable 2817 allocations. If both kernelcore and movablecore is 2818 specified, then kernelcore will be at *least* the 2819 specified value but may be more. If movablecore on its 2820 own is specified, the administrator must be careful 2821 that the amount of memory usable for all allocations 2822 is not too small. 2823 2824 movable_node [KNL] Boot-time switch to make hotplugable memory 2825 NUMA nodes to be movable. This means that the memory 2826 of such nodes will be usable only for movable 2827 allocations which rules out almost all kernel 2828 allocations. Use with caution! 2829 2830 MTD_Partition= [MTD] 2831 Format: <name>,<region-number>,<size>,<offset> 2832 2833 MTD_Region= [MTD] Format: 2834 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>] 2835 2836 mtdparts= [MTD] 2837 See drivers/mtd/parsers/cmdlinepart.c 2838 2839 multitce=off [PPC] This parameter disables the use of the pSeries 2840 firmware feature for updating multiple TCE entries 2841 at a time. 2842 2843 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration 2844 2845 Format: [die0_boundary][,die0_lock][,die1_boundary][,die1_lock] 2846 2847 boundary - index of last SLC block on Flex-OneNAND. 2848 The remaining blocks are configured as MLC blocks. 2849 lock - Configure if Flex-OneNAND boundary should be locked. 2850 Once locked, the boundary cannot be changed. 2851 1 indicates lock status, 0 indicates unlock status. 2852 2853 mtdset= [ARM] 2854 ARM/S3C2412 JIVE boot control 2855 2856 See arch/arm/mach-s3c2412/mach-jive.c 2857 2858 mtouchusb.raw_coordinates= 2859 [HW] Make the MicroTouch USB driver use raw coordinates 2860 ('y', default) or cooked coordinates ('n') 2861 2862 mtrr_chunk_size=nn[KMG] [X86] 2863 used for mtrr cleanup. It is largest continuous chunk 2864 that could hold holes aka. UC entries. 2865 2866 mtrr_gran_size=nn[KMG] [X86] 2867 Used for mtrr cleanup. It is granularity of mtrr block. 2868 Default is 1. 2869 Large value could prevent small alignment from 2870 using up MTRRs. 2871 2872 mtrr_spare_reg_nr=n [X86] 2873 Format: <integer> 2874 Range: 0,7 : spare reg number 2875 Default : 1 2876 Used for mtrr cleanup. It is spare mtrr entries number. 2877 Set to 2 or more if your graphical card needs more. 2878 2879 n2= [NET] SDL Inc. RISCom/N2 synchronous serial card 2880 2881 netdev= [NET] Network devices parameters 2882 Format: <irq>,<io>,<mem_start>,<mem_end>,<name> 2883 Note that mem_start is often overloaded to mean 2884 something different and driver-specific. 2885 This usage is only documented in each driver source 2886 file if at all. 2887 2888 nf_conntrack.acct= 2889 [NETFILTER] Enable connection tracking flow accounting 2890 0 to disable accounting 2891 1 to enable accounting 2892 Default value is 0. 2893 2894 nfsaddrs= [NFS] Deprecated. Use ip= instead. 2895 See Documentation/admin-guide/nfs/nfsroot.rst. 2896 2897 nfsroot= [NFS] nfs root filesystem for disk-less boxes. 2898 See Documentation/admin-guide/nfs/nfsroot.rst. 2899 2900 nfsrootdebug [NFS] enable nfsroot debugging messages. 2901 See Documentation/admin-guide/nfs/nfsroot.rst. 2902 2903 nfs.callback_nr_threads= 2904 [NFSv4] set the total number of threads that the 2905 NFS client will assign to service NFSv4 callback 2906 requests. 2907 2908 nfs.callback_tcpport= 2909 [NFS] set the TCP port on which the NFSv4 callback 2910 channel should listen. 2911 2912 nfs.cache_getent= 2913 [NFS] sets the pathname to the program which is used 2914 to update the NFS client cache entries. 2915 2916 nfs.cache_getent_timeout= 2917 [NFS] sets the timeout after which an attempt to 2918 update a cache entry is deemed to have failed. 2919 2920 nfs.idmap_cache_timeout= 2921 [NFS] set the maximum lifetime for idmapper cache 2922 entries. 2923 2924 nfs.enable_ino64= 2925 [NFS] enable 64-bit inode numbers. 2926 If zero, the NFS client will fake up a 32-bit inode 2927 number for the readdir() and stat() syscalls instead 2928 of returning the full 64-bit number. 2929 The default is to return 64-bit inode numbers. 2930 2931 nfs.max_session_cb_slots= 2932 [NFSv4.1] Sets the maximum number of session 2933 slots the client will assign to the callback 2934 channel. This determines the maximum number of 2935 callbacks the client will process in parallel for 2936 a particular server. 2937 2938 nfs.max_session_slots= 2939 [NFSv4.1] Sets the maximum number of session slots 2940 the client will attempt to negotiate with the server. 2941 This limits the number of simultaneous RPC requests 2942 that the client can send to the NFSv4.1 server. 2943 Note that there is little point in setting this 2944 value higher than the max_tcp_slot_table_limit. 2945 2946 nfs.nfs4_disable_idmapping= 2947 [NFSv4] When set to the default of '1', this option 2948 ensures that both the RPC level authentication 2949 scheme and the NFS level operations agree to use 2950 numeric uids/gids if the mount is using the 2951 'sec=sys' security flavour. In effect it is 2952 disabling idmapping, which can make migration from 2953 legacy NFSv2/v3 systems to NFSv4 easier. 2954 Servers that do not support this mode of operation 2955 will be autodetected by the client, and it will fall 2956 back to using the idmapper. 2957 To turn off this behaviour, set the value to '0'. 2958 nfs.nfs4_unique_id= 2959 [NFS4] Specify an additional fixed unique ident- 2960 ification string that NFSv4 clients can insert into 2961 their nfs_client_id4 string. This is typically a 2962 UUID that is generated at system install time. 2963 2964 nfs.send_implementation_id = 2965 [NFSv4.1] Send client implementation identification 2966 information in exchange_id requests. 2967 If zero, no implementation identification information 2968 will be sent. 2969 The default is to send the implementation identification 2970 information. 2971 2972 nfs.recover_lost_locks = 2973 [NFSv4] Attempt to recover locks that were lost due 2974 to a lease timeout on the server. Please note that 2975 doing this risks data corruption, since there are 2976 no guarantees that the file will remain unchanged 2977 after the locks are lost. 2978 If you want to enable the kernel legacy behaviour of 2979 attempting to recover these locks, then set this 2980 parameter to '1'. 2981 The default parameter value of '0' causes the kernel 2982 not to attempt recovery of lost locks. 2983 2984 nfs4.layoutstats_timer = 2985 [NFSv4.2] Change the rate at which the kernel sends 2986 layoutstats to the pNFS metadata server. 2987 2988 Setting this to value to 0 causes the kernel to use 2989 whatever value is the default set by the layout 2990 driver. A non-zero value sets the minimum interval 2991 in seconds between layoutstats transmissions. 2992 2993 nfsd.nfs4_disable_idmapping= 2994 [NFSv4] When set to the default of '1', the NFSv4 2995 server will return only numeric uids and gids to 2996 clients using auth_sys, and will accept numeric uids 2997 and gids from such clients. This is intended to ease 2998 migration from NFSv2/v3. 2999 3000 nmi_debug= [KNL,SH] Specify one or more actions to take 3001 when a NMI is triggered. 3002 Format: [state][,regs][,debounce][,die] 3003 3004 nmi_watchdog= [KNL,BUGS=X86] Debugging features for SMP kernels 3005 Format: [panic,][nopanic,][num] 3006 Valid num: 0 or 1 3007 0 - turn hardlockup detector in nmi_watchdog off 3008 1 - turn hardlockup detector in nmi_watchdog on 3009 When panic is specified, panic when an NMI watchdog 3010 timeout occurs (or 'nopanic' to not panic on an NMI 3011 watchdog, if CONFIG_BOOTPARAM_HARDLOCKUP_PANIC is set) 3012 To disable both hard and soft lockup detectors, 3013 please see 'nowatchdog'. 3014 This is useful when you use a panic=... timeout and 3015 need the box quickly up again. 3016 3017 These settings can be accessed at runtime via 3018 the nmi_watchdog and hardlockup_panic sysctls. 3019 3020 netpoll.carrier_timeout= 3021 [NET] Specifies amount of time (in seconds) that 3022 netpoll should wait for a carrier. By default netpoll 3023 waits 4 seconds. 3024 3025 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths 3026 emulation library even if a 387 maths coprocessor 3027 is present. 3028 3029 no5lvl [X86-64] Disable 5-level paging mode. Forces 3030 kernel to use 4-level paging instead. 3031 3032 no_console_suspend 3033 [HW] Never suspend the console 3034 Disable suspending of consoles during suspend and 3035 hibernate operations. Once disabled, debugging 3036 messages can reach various consoles while the rest 3037 of the system is being put to sleep (ie, while 3038 debugging driver suspend/resume hooks). This may 3039 not work reliably with all consoles, but is known 3040 to work with serial and VGA consoles. 3041 To facilitate more flexible debugging, we also add 3042 console_suspend, a printk module parameter to control 3043 it. Users could use console_suspend (usually 3044 /sys/module/printk/parameters/console_suspend) to 3045 turn on/off it dynamically. 3046 3047 novmcoredd [KNL,KDUMP] 3048 Disable device dump. Device dump allows drivers to 3049 append dump data to vmcore so you can collect driver 3050 specified debug info. Drivers can append the data 3051 without any limit and this data is stored in memory, 3052 so this may cause significant memory stress. Disabling 3053 device dump can help save memory but the driver debug 3054 data will be no longer available. This parameter 3055 is only available when CONFIG_PROC_VMCORE_DEVICE_DUMP 3056 is set. 3057 3058 noaliencache [MM, NUMA, SLAB] Disables the allocation of alien 3059 caches in the slab allocator. Saves per-node memory, 3060 but will impact performance. 3061 3062 noalign [KNL,ARM] 3063 3064 noaltinstr [S390] Disables alternative instructions patching 3065 (CPU alternatives feature). 3066 3067 noapic [SMP,APIC] Tells the kernel to not make use of any 3068 IOAPICs that may be present in the system. 3069 3070 noautogroup Disable scheduler automatic task group creation. 3071 3072 nobats [PPC] Do not use BATs for mapping kernel lowmem 3073 on "Classic" PPC cores. 3074 3075 nocache [ARM] 3076 3077 noclflush [BUGS=X86] Don't use the CLFLUSH instruction 3078 3079 nodelayacct [KNL] Disable per-task delay accounting 3080 3081 nodsp [SH] Disable hardware DSP at boot time. 3082 3083 noefi Disable EFI runtime services support. 3084 3085 noexec [IA-64] 3086 3087 noexec [X86] 3088 On X86-32 available only on PAE configured kernels. 3089 noexec=on: enable non-executable mappings (default) 3090 noexec=off: disable non-executable mappings 3091 3092 nosmap [X86,PPC] 3093 Disable SMAP (Supervisor Mode Access Prevention) 3094 even if it is supported by processor. 3095 3096 nosmep [X86,PPC] 3097 Disable SMEP (Supervisor Mode Execution Prevention) 3098 even if it is supported by processor. 3099 3100 noexec32 [X86-64] 3101 This affects only 32-bit executables. 3102 noexec32=on: enable non-executable mappings (default) 3103 read doesn't imply executable mappings 3104 noexec32=off: disable non-executable mappings 3105 read implies executable mappings 3106 3107 nofpu [MIPS,SH] Disable hardware FPU at boot time. 3108 3109 nofxsr [BUGS=X86-32] Disables x86 floating point extended 3110 register save and restore. The kernel will only save 3111 legacy floating-point registers on task switch. 3112 3113 nohugeiomap [KNL,x86,PPC] Disable kernel huge I/O mappings. 3114 3115 nosmt [KNL,S390] Disable symmetric multithreading (SMT). 3116 Equivalent to smt=1. 3117 3118 [KNL,x86] Disable symmetric multithreading (SMT). 3119 nosmt=force: Force disable SMT, cannot be undone 3120 via the sysfs control file. 3121 3122 nospectre_v1 [X86,PPC] Disable mitigations for Spectre Variant 1 3123 (bounds check bypass). With this option data leaks are 3124 possible in the system. 3125 3126 nospectre_v2 [X86,PPC_FSL_BOOK3E,ARM64] Disable all mitigations for 3127 the Spectre variant 2 (indirect branch prediction) 3128 vulnerability. System may allow data leaks with this 3129 option. 3130 3131 nospec_store_bypass_disable 3132 [HW] Disable all mitigations for the Speculative Store Bypass vulnerability 3133 3134 noxsave [BUGS=X86] Disables x86 extended register state save 3135 and restore using xsave. The kernel will fallback to 3136 enabling legacy floating-point and sse state. 3137 3138 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended 3139 register states. The kernel will fall back to use 3140 xsave to save the states. By using this parameter, 3141 performance of saving the states is degraded because 3142 xsave doesn't support modified optimization while 3143 xsaveopt supports it on xsaveopt enabled systems. 3144 3145 noxsaves [X86] Disables xsaves and xrstors used in saving and 3146 restoring x86 extended register state in compacted 3147 form of xsave area. The kernel will fall back to use 3148 xsaveopt and xrstor to save and restore the states 3149 in standard form of xsave area. By using this 3150 parameter, xsave area per process might occupy more 3151 memory on xsaves enabled systems. 3152 3153 nohlt [BUGS=ARM,SH] Tells the kernel that the sleep(SH) or 3154 wfi(ARM) instruction doesn't work correctly and not to 3155 use it. This is also useful when using JTAG debugger. 3156 3157 no_file_caps Tells the kernel not to honor file capabilities. The 3158 only way then for a file to be executed with privilege 3159 is to be setuid root or executed by root. 3160 3161 nohalt [IA-64] Tells the kernel not to use the power saving 3162 function PAL_HALT_LIGHT when idle. This increases 3163 power-consumption. On the positive side, it reduces 3164 interrupt wake-up latency, which may improve performance 3165 in certain environments such as networked servers or 3166 real-time systems. 3167 3168 nohibernate [HIBERNATION] Disable hibernation and resume. 3169 3170 nohz= [KNL] Boottime enable/disable dynamic ticks 3171 Valid arguments: on, off 3172 Default: on 3173 3174 nohz_full= [KNL,BOOT,SMP,ISOL] 3175 The argument is a cpu list, as described above. 3176 In kernels built with CONFIG_NO_HZ_FULL=y, set 3177 the specified list of CPUs whose tick will be stopped 3178 whenever possible. The boot CPU will be forced outside 3179 the range to maintain the timekeeping. Any CPUs 3180 in this list will have their RCU callbacks offloaded, 3181 just as if they had also been called out in the 3182 rcu_nocbs= boot parameter. 3183 3184 noiotrap [SH] Disables trapped I/O port accesses. 3185 3186 noirqdebug [X86-32] Disables the code which attempts to detect and 3187 disable unhandled interrupt sources. 3188 3189 no_timer_check [X86,APIC] Disables the code which tests for 3190 broken timer IRQ sources. 3191 3192 noisapnp [ISAPNP] Disables ISA PnP code. 3193 3194 noinitrd [RAM] Tells the kernel not to load any configured 3195 initial RAM disk. 3196 3197 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt 3198 remapping. 3199 [Deprecated - use intremap=off] 3200 3201 nointroute [IA-64] 3202 3203 noinvpcid [X86] Disable the INVPCID cpu feature. 3204 3205 nojitter [IA-64] Disables jitter checking for ITC timers. 3206 3207 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver 3208 3209 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page 3210 fault handling. 3211 3212 no-vmw-sched-clock 3213 [X86,PV_OPS] Disable paravirtualized VMware scheduler 3214 clock and use the default one. 3215 3216 no-steal-acc [X86,PV_OPS,ARM64] Disable paravirtualized steal time 3217 accounting. steal time is computed, but won't 3218 influence scheduler behaviour 3219 3220 nolapic [X86-32,APIC] Do not enable or use the local APIC. 3221 3222 nolapic_timer [X86-32,APIC] Do not use the local APIC timer. 3223 3224 noltlbs [PPC] Do not use large page/tlb entries for kernel 3225 lowmem mapping on PPC40x and PPC8xx 3226 3227 nomca [IA-64] Disable machine check abort handling 3228 3229 nomce [X86-32] Disable Machine Check Exception 3230 3231 nomfgpt [X86-32] Disable Multi-Function General Purpose 3232 Timer usage (for AMD Geode machines). 3233 3234 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to 3235 shutdown the other cpus. Instead use the REBOOT_VECTOR 3236 irq. 3237 3238 nomodule Disable module load 3239 3240 nopat [X86] Disable PAT (page attribute table extension of 3241 pagetables) support. 3242 3243 nopcid [X86-64] Disable the PCID cpu feature. 3244 3245 norandmaps Don't use address space randomization. Equivalent to 3246 echo 0 > /proc/sys/kernel/randomize_va_space 3247 3248 noreplace-smp [X86-32,SMP] Don't replace SMP instructions 3249 with UP alternatives 3250 3251 nordrand [X86] Disable kernel use of the RDRAND and 3252 RDSEED instructions even if they are supported 3253 by the processor. RDRAND and RDSEED are still 3254 available to user space applications. 3255 3256 noresume [SWSUSP] Disables resume and restores original swap 3257 space. 3258 3259 no-scroll [VGA] Disables scrollback. 3260 This is required for the Braillex ib80-piezo Braille 3261 reader made by F.H. Papenmeier (Germany). 3262 3263 nosbagart [IA-64] 3264 3265 nosep [BUGS=X86-32] Disables x86 SYSENTER/SYSEXIT support. 3266 3267 nosmp [SMP] Tells an SMP kernel to act as a UP kernel, 3268 and disable the IO APIC. legacy for "maxcpus=0". 3269 3270 nosoftlockup [KNL] Disable the soft-lockup detector. 3271 3272 nosync [HW,M68K] Disables sync negotiation for all devices. 3273 3274 nowatchdog [KNL] Disable both lockup detectors, i.e. 3275 soft-lockup and NMI watchdog (hard-lockup). 3276 3277 nowb [ARM] 3278 3279 nox2apic [X86-64,APIC] Do not enable x2APIC mode. 3280 3281 cpu0_hotplug [X86] Turn on CPU0 hotplug feature when 3282 CONFIG_BOOTPARAM_HOTPLUG_CPU0 is off. 3283 Some features depend on CPU0. Known dependencies are: 3284 1. Resume from suspend/hibernate depends on CPU0. 3285 Suspend/hibernate will fail if CPU0 is offline and you 3286 need to online CPU0 before suspend/hibernate. 3287 2. PIC interrupts also depend on CPU0. CPU0 can't be 3288 removed if a PIC interrupt is detected. 3289 It's said poweroff/reboot may depend on CPU0 on some 3290 machines although I haven't seen such issues so far 3291 after CPU0 is offline on a few tested machines. 3292 If the dependencies are under your control, you can 3293 turn on cpu0_hotplug. 3294 3295 nps_mtm_hs_ctr= [KNL,ARC] 3296 This parameter sets the maximum duration, in 3297 cycles, each HW thread of the CTOP can run 3298 without interruptions, before HW switches it. 3299 The actual maximum duration is 16 times this 3300 parameter's value. 3301 Format: integer between 1 and 255 3302 Default: 255 3303 3304 nptcg= [IA-64] Override max number of concurrent global TLB 3305 purges which is reported from either PAL_VM_SUMMARY or 3306 SAL PALO. 3307 3308 nr_cpus= [SMP] Maximum number of processors that an SMP kernel 3309 could support. nr_cpus=n : n >= 1 limits the kernel to 3310 support 'n' processors. It could be larger than the 3311 number of already plugged CPU during bootup, later in 3312 runtime you can physically add extra cpu until it reaches 3313 n. So during boot up some boot time memory for per-cpu 3314 variables need be pre-allocated for later physical cpu 3315 hot plugging. 3316 3317 nr_uarts= [SERIAL] maximum number of UARTs to be registered. 3318 3319 numa_balancing= [KNL,X86] Enable or disable automatic NUMA balancing. 3320 Allowed values are enable and disable 3321 3322 numa_zonelist_order= [KNL, BOOT] Select zonelist order for NUMA. 3323 'node', 'default' can be specified 3324 This can be set from sysctl after boot. 3325 See Documentation/admin-guide/sysctl/vm.rst for details. 3326 3327 ohci1394_dma=early [HW] enable debugging via the ohci1394 driver. 3328 See Documentation/debugging-via-ohci1394.txt for more 3329 info. 3330 3331 olpc_ec_timeout= [OLPC] ms delay when issuing EC commands 3332 Rather than timing out after 20 ms if an EC 3333 command is not properly ACKed, override the length 3334 of the timeout. We have interrupts disabled while 3335 waiting for the ACK, so if this is set too high 3336 interrupts *may* be lost! 3337 3338 omap_mux= [OMAP] Override bootloader pin multiplexing. 3339 Format: <mux_mode0.mode_name=value>... 3340 For example, to override I2C bus2: 3341 omap_mux=i2c2_scl.i2c2_scl=0x100,i2c2_sda.i2c2_sda=0x100 3342 3343 oprofile.timer= [HW] 3344 Use timer interrupt instead of performance counters 3345 3346 oprofile.cpu_type= Force an oprofile cpu type 3347 This might be useful if you have an older oprofile 3348 userland or if you want common events. 3349 Format: { arch_perfmon } 3350 arch_perfmon: [X86] Force use of architectural 3351 perfmon on Intel CPUs instead of the 3352 CPU specific event set. 3353 timer: [X86] Force use of architectural NMI 3354 timer mode (see also oprofile.timer 3355 for generic hr timer mode) 3356 3357 oops=panic Always panic on oopses. Default is to just kill the 3358 process, but there is a small probability of 3359 deadlocking the machine. 3360 This will also cause panics on machine check exceptions. 3361 Useful together with panic=30 to trigger a reboot. 3362 3363 page_alloc.shuffle= 3364 [KNL] Boolean flag to control whether the page allocator 3365 should randomize its free lists. The randomization may 3366 be automatically enabled if the kernel detects it is 3367 running on a platform with a direct-mapped memory-side 3368 cache, and this parameter can be used to 3369 override/disable that behavior. The state of the flag 3370 can be read from sysfs at: 3371 /sys/module/page_alloc/parameters/shuffle. 3372 3373 page_owner= [KNL] Boot-time page_owner enabling option. 3374 Storage of the information about who allocated 3375 each page is disabled in default. With this switch, 3376 we can turn it on. 3377 on: enable the feature 3378 3379 page_poison= [KNL] Boot-time parameter changing the state of 3380 poisoning on the buddy allocator, available with 3381 CONFIG_PAGE_POISONING=y. 3382 off: turn off poisoning (default) 3383 on: turn on poisoning 3384 3385 panic= [KNL] Kernel behaviour on panic: delay <timeout> 3386 timeout > 0: seconds before rebooting 3387 timeout = 0: wait forever 3388 timeout < 0: reboot immediately 3389 Format: <timeout> 3390 3391 panic_print= Bitmask for printing system info when panic happens. 3392 User can chose combination of the following bits: 3393 bit 0: print all tasks info 3394 bit 1: print system memory info 3395 bit 2: print timer info 3396 bit 3: print locks info if CONFIG_LOCKDEP is on 3397 bit 4: print ftrace buffer 3398 bit 5: print all printk messages in buffer 3399 3400 panic_on_warn panic() instead of WARN(). Useful to cause kdump 3401 on a WARN(). 3402 3403 crash_kexec_post_notifiers 3404 Run kdump after running panic-notifiers and dumping 3405 kmsg. This only for the users who doubt kdump always 3406 succeeds in any situation. 3407 Note that this also increases risks of kdump failure, 3408 because some panic notifiers can make the crashed 3409 kernel more unstable. 3410 3411 parkbd.port= [HW] Parallel port number the keyboard adapter is 3412 connected to, default is 0. 3413 Format: <parport#> 3414 parkbd.mode= [HW] Parallel port keyboard adapter mode of operation, 3415 0 for XT, 1 for AT (default is AT). 3416 Format: <mode> 3417 3418 parport= [HW,PPT] Specify parallel ports. 0 disables. 3419 Format: { 0 | auto | 0xBBB[,IRQ[,DMA]] } 3420 Use 'auto' to force the driver to use any 3421 IRQ/DMA settings detected (the default is to 3422 ignore detected IRQ/DMA settings because of 3423 possible conflicts). You can specify the base 3424 address, IRQ, and DMA settings; IRQ and DMA 3425 should be numbers, or 'auto' (for using detected 3426 settings on that particular port), or 'nofifo' 3427 (to avoid using a FIFO even if it is detected). 3428 Parallel ports are assigned in the order they 3429 are specified on the command line, starting 3430 with parport0. 3431 3432 parport_init_mode= [HW,PPT] 3433 Configure VIA parallel port to operate in 3434 a specific mode. This is necessary on Pegasos 3435 computer where firmware has no options for setting 3436 up parallel port mode and sets it to spp. 3437 Currently this function knows 686a and 8231 chips. 3438 Format: [spp|ps2|epp|ecp|ecpepp] 3439 3440 pause_on_oops= 3441 Halt all CPUs after the first oops has been printed for 3442 the specified number of seconds. This is to be used if 3443 your oopses keep scrolling off the screen. 3444 3445 pcbit= [HW,ISDN] 3446 3447 pcd. [PARIDE] 3448 See header of drivers/block/paride/pcd.c. 3449 See also Documentation/admin-guide/blockdev/paride.rst. 3450 3451 pci=option[,option...] [PCI] various PCI subsystem options. 3452 3453 Some options herein operate on a specific device 3454 or a set of devices (<pci_dev>). These are 3455 specified in one of the following formats: 3456 3457 [<domain>:]<bus>:<dev>.<func>[/<dev>.<func>]* 3458 pci:<vendor>:<device>[:<subvendor>:<subdevice>] 3459 3460 Note: the first format specifies a PCI 3461 bus/device/function address which may change 3462 if new hardware is inserted, if motherboard 3463 firmware changes, or due to changes caused 3464 by other kernel parameters. If the 3465 domain is left unspecified, it is 3466 taken to be zero. Optionally, a path 3467 to a device through multiple device/function 3468 addresses can be specified after the base 3469 address (this is more robust against 3470 renumbering issues). The second format 3471 selects devices using IDs from the 3472 configuration space which may match multiple 3473 devices in the system. 3474 3475 earlydump dump PCI config space before the kernel 3476 changes anything 3477 off [X86] don't probe for the PCI bus 3478 bios [X86-32] force use of PCI BIOS, don't access 3479 the hardware directly. Use this if your machine 3480 has a non-standard PCI host bridge. 3481 nobios [X86-32] disallow use of PCI BIOS, only direct 3482 hardware access methods are allowed. Use this 3483 if you experience crashes upon bootup and you 3484 suspect they are caused by the BIOS. 3485 conf1 [X86] Force use of PCI Configuration Access 3486 Mechanism 1 (config address in IO port 0xCF8, 3487 data in IO port 0xCFC, both 32-bit). 3488 conf2 [X86] Force use of PCI Configuration Access 3489 Mechanism 2 (IO port 0xCF8 is an 8-bit port for 3490 the function, IO port 0xCFA, also 8-bit, sets 3491 bus number. The config space is then accessed 3492 through ports 0xC000-0xCFFF). 3493 See http://wiki.osdev.org/PCI for more info 3494 on the configuration access mechanisms. 3495 noaer [PCIE] If the PCIEAER kernel config parameter is 3496 enabled, this kernel boot option can be used to 3497 disable the use of PCIE advanced error reporting. 3498 nodomains [PCI] Disable support for multiple PCI 3499 root domains (aka PCI segments, in ACPI-speak). 3500 nommconf [X86] Disable use of MMCONFIG for PCI 3501 Configuration 3502 check_enable_amd_mmconf [X86] check for and enable 3503 properly configured MMIO access to PCI 3504 config space on AMD family 10h CPU 3505 nomsi [MSI] If the PCI_MSI kernel config parameter is 3506 enabled, this kernel boot option can be used to 3507 disable the use of MSI interrupts system-wide. 3508 noioapicquirk [APIC] Disable all boot interrupt quirks. 3509 Safety option to keep boot IRQs enabled. This 3510 should never be necessary. 3511 ioapicreroute [APIC] Enable rerouting of boot IRQs to the 3512 primary IO-APIC for bridges that cannot disable 3513 boot IRQs. This fixes a source of spurious IRQs 3514 when the system masks IRQs. 3515 noioapicreroute [APIC] Disable workaround that uses the 3516 boot IRQ equivalent of an IRQ that connects to 3517 a chipset where boot IRQs cannot be disabled. 3518 The opposite of ioapicreroute. 3519 biosirq [X86-32] Use PCI BIOS calls to get the interrupt 3520 routing table. These calls are known to be buggy 3521 on several machines and they hang the machine 3522 when used, but on other computers it's the only 3523 way to get the interrupt routing table. Try 3524 this option if the kernel is unable to allocate 3525 IRQs or discover secondary PCI buses on your 3526 motherboard. 3527 rom [X86] Assign address space to expansion ROMs. 3528 Use with caution as certain devices share 3529 address decoders between ROMs and other 3530 resources. 3531 norom [X86] Do not assign address space to 3532 expansion ROMs that do not already have 3533 BIOS assigned address ranges. 3534 nobar [X86] Do not assign address space to the 3535 BARs that weren't assigned by the BIOS. 3536 irqmask=0xMMMM [X86] Set a bit mask of IRQs allowed to be 3537 assigned automatically to PCI devices. You can 3538 make the kernel exclude IRQs of your ISA cards 3539 this way. 3540 pirqaddr=0xAAAAA [X86] Specify the physical address 3541 of the PIRQ table (normally generated 3542 by the BIOS) if it is outside the 3543 F0000h-100000h range. 3544 lastbus=N [X86] Scan all buses thru bus #N. Can be 3545 useful if the kernel is unable to find your 3546 secondary buses and you want to tell it 3547 explicitly which ones they are. 3548 assign-busses [X86] Always assign all PCI bus 3549 numbers ourselves, overriding 3550 whatever the firmware may have done. 3551 usepirqmask [X86] Honor the possible IRQ mask stored 3552 in the BIOS $PIR table. This is needed on 3553 some systems with broken BIOSes, notably 3554 some HP Pavilion N5400 and Omnibook XE3 3555 notebooks. This will have no effect if ACPI 3556 IRQ routing is enabled. 3557 noacpi [X86] Do not use ACPI for IRQ routing 3558 or for PCI scanning. 3559 use_crs [X86] Use PCI host bridge window information 3560 from ACPI. On BIOSes from 2008 or later, this 3561 is enabled by default. If you need to use this, 3562 please report a bug. 3563 nocrs [X86] Ignore PCI host bridge windows from ACPI. 3564 If you need to use this, please report a bug. 3565 routeirq Do IRQ routing for all PCI devices. 3566 This is normally done in pci_enable_device(), 3567 so this option is a temporary workaround 3568 for broken drivers that don't call it. 3569 skip_isa_align [X86] do not align io start addr, so can 3570 handle more pci cards 3571 noearly [X86] Don't do any early type 1 scanning. 3572 This might help on some broken boards which 3573 machine check when some devices' config space 3574 is read. But various workarounds are disabled 3575 and some IOMMU drivers will not work. 3576 bfsort Sort PCI devices into breadth-first order. 3577 This sorting is done to get a device 3578 order compatible with older (<= 2.4) kernels. 3579 nobfsort Don't sort PCI devices into breadth-first order. 3580 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size) 3581 tuning and use the BIOS-configured MPS defaults. 3582 pcie_bus_safe Set every device's MPS to the largest value 3583 supported by all devices below the root complex. 3584 pcie_bus_perf Set device MPS to the largest allowable MPS 3585 based on its parent bus. Also set MRRS (Max 3586 Read Request Size) to the largest supported 3587 value (no larger than the MPS that the device 3588 or bus can support) for best performance. 3589 pcie_bus_peer2peer Set every device's MPS to 128B, which 3590 every device is guaranteed to support. This 3591 configuration allows peer-to-peer DMA between 3592 any pair of devices, possibly at the cost of 3593 reduced performance. This also guarantees 3594 that hot-added devices will work. 3595 cbiosize=nn[KMG] The fixed amount of bus space which is 3596 reserved for the CardBus bridge's IO window. 3597 The default value is 256 bytes. 3598 cbmemsize=nn[KMG] The fixed amount of bus space which is 3599 reserved for the CardBus bridge's memory 3600 window. The default value is 64 megabytes. 3601 resource_alignment= 3602 Format: 3603 [<order of align>@]<pci_dev>[; ...] 3604 Specifies alignment and device to reassign 3605 aligned memory resources. How to 3606 specify the device is described above. 3607 If <order of align> is not specified, 3608 PAGE_SIZE is used as alignment. 3609 A PCI-PCI bridge can be specified if resource 3610 windows need to be expanded. 3611 To specify the alignment for several 3612 instances of a device, the PCI vendor, 3613 device, subvendor, and subdevice may be 3614 specified, e.g., 12@pci:8086:9c22:103c:198f 3615 for 4096-byte alignment. 3616 ecrc= Enable/disable PCIe ECRC (transaction layer 3617 end-to-end CRC checking). 3618 bios: Use BIOS/firmware settings. This is the 3619 the default. 3620 off: Turn ECRC off 3621 on: Turn ECRC on. 3622 hpiosize=nn[KMG] The fixed amount of bus space which is 3623 reserved for hotplug bridge's IO window. 3624 Default size is 256 bytes. 3625 hpmmiosize=nn[KMG] The fixed amount of bus space which is 3626 reserved for hotplug bridge's MMIO window. 3627 Default size is 2 megabytes. 3628 hpmmioprefsize=nn[KMG] The fixed amount of bus space which is 3629 reserved for hotplug bridge's MMIO_PREF window. 3630 Default size is 2 megabytes. 3631 hpmemsize=nn[KMG] The fixed amount of bus space which is 3632 reserved for hotplug bridge's MMIO and 3633 MMIO_PREF window. 3634 Default size is 2 megabytes. 3635 hpbussize=nn The minimum amount of additional bus numbers 3636 reserved for buses below a hotplug bridge. 3637 Default is 1. 3638 realloc= Enable/disable reallocating PCI bridge resources 3639 if allocations done by BIOS are too small to 3640 accommodate resources required by all child 3641 devices. 3642 off: Turn realloc off 3643 on: Turn realloc on 3644 realloc same as realloc=on 3645 noari do not use PCIe ARI. 3646 noats [PCIE, Intel-IOMMU, AMD-IOMMU] 3647 do not use PCIe ATS (and IOMMU device IOTLB). 3648 pcie_scan_all Scan all possible PCIe devices. Otherwise we 3649 only look for one device below a PCIe downstream 3650 port. 3651 big_root_window Try to add a big 64bit memory window to the PCIe 3652 root complex on AMD CPUs. Some GFX hardware 3653 can resize a BAR to allow access to all VRAM. 3654 Adding the window is slightly risky (it may 3655 conflict with unreported devices), so this 3656 taints the kernel. 3657 disable_acs_redir=<pci_dev>[; ...] 3658 Specify one or more PCI devices (in the format 3659 specified above) separated by semicolons. 3660 Each device specified will have the PCI ACS 3661 redirect capabilities forced off which will 3662 allow P2P traffic between devices through 3663 bridges without forcing it upstream. Note: 3664 this removes isolation between devices and 3665 may put more devices in an IOMMU group. 3666 force_floating [S390] Force usage of floating interrupts. 3667 nomio [S390] Do not use MIO instructions. 3668 3669 pcie_aspm= [PCIE] Forcibly enable or disable PCIe Active State Power 3670 Management. 3671 off Disable ASPM. 3672 force Enable ASPM even on devices that claim not to support it. 3673 WARNING: Forcing ASPM on may cause system lockups. 3674 3675 pcie_ports= [PCIE] PCIe port services handling: 3676 native Use native PCIe services (PME, AER, DPC, PCIe hotplug) 3677 even if the platform doesn't give the OS permission to 3678 use them. This may cause conflicts if the platform 3679 also tries to use these services. 3680 dpc-native Use native PCIe service for DPC only. May 3681 cause conflicts if firmware uses AER or DPC. 3682 compat Disable native PCIe services (PME, AER, DPC, PCIe 3683 hotplug). 3684 3685 pcie_port_pm= [PCIE] PCIe port power management handling: 3686 off Disable power management of all PCIe ports 3687 force Forcibly enable power management of all PCIe ports 3688 3689 pcie_pme= [PCIE,PM] Native PCIe PME signaling options: 3690 nomsi Do not use MSI for native PCIe PME signaling (this makes 3691 all PCIe root ports use INTx for all services). 3692 3693 pcmv= [HW,PCMCIA] BadgePAD 4 3694 3695 pd_ignore_unused 3696 [PM] 3697 Keep all power-domains already enabled by bootloader on, 3698 even if no driver has claimed them. This is useful 3699 for debug and development, but should not be 3700 needed on a platform with proper driver support. 3701 3702 pd. [PARIDE] 3703 See Documentation/admin-guide/blockdev/paride.rst. 3704 3705 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at 3706 boot time. 3707 Format: { 0 | 1 } 3708 See arch/parisc/kernel/pdc_chassis.c 3709 3710 percpu_alloc= Select which percpu first chunk allocator to use. 3711 Currently supported values are "embed" and "page". 3712 Archs may support subset or none of the selections. 3713 See comments in mm/percpu.c for details on each 3714 allocator. This parameter is primarily for debugging 3715 and performance comparison. 3716 3717 pf. [PARIDE] 3718 See Documentation/admin-guide/blockdev/paride.rst. 3719 3720 pg. [PARIDE] 3721 See Documentation/admin-guide/blockdev/paride.rst. 3722 3723 pirq= [SMP,APIC] Manual mp-table setup 3724 See Documentation/x86/i386/IO-APIC.rst. 3725 3726 plip= [PPT,NET] Parallel port network link 3727 Format: { parport<nr> | timid | 0 } 3728 See also Documentation/admin-guide/parport.rst. 3729 3730 pmtmr= [X86] Manual setup of pmtmr I/O Port. 3731 Override pmtimer IOPort with a hex value. 3732 e.g. pmtmr=0x508 3733 3734 pm_debug_messages [SUSPEND,KNL] 3735 Enable suspend/resume debug messages during boot up. 3736 3737 pnp.debug=1 [PNP] 3738 Enable PNP debug messages (depends on the 3739 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time 3740 via /sys/module/pnp/parameters/debug. We always show 3741 current resource usage; turning this on also shows 3742 possible settings and some assignment information. 3743 3744 pnpacpi= [ACPI] 3745 { off } 3746 3747 pnpbios= [ISAPNP] 3748 { on | off | curr | res | no-curr | no-res } 3749 3750 pnp_reserve_irq= 3751 [ISAPNP] Exclude IRQs for the autoconfiguration 3752 3753 pnp_reserve_dma= 3754 [ISAPNP] Exclude DMAs for the autoconfiguration 3755 3756 pnp_reserve_io= [ISAPNP] Exclude I/O ports for the autoconfiguration 3757 Ranges are in pairs (I/O port base and size). 3758 3759 pnp_reserve_mem= 3760 [ISAPNP] Exclude memory regions for the 3761 autoconfiguration. 3762 Ranges are in pairs (memory base and size). 3763 3764 ports= [IP_VS_FTP] IPVS ftp helper module 3765 Default is 21. 3766 Up to 8 (IP_VS_APP_MAX_PORTS) ports 3767 may be specified. 3768 Format: <port>,<port>.... 3769 3770 powersave=off [PPC] This option disables power saving features. 3771 It specifically disables cpuidle and sets the 3772 platform machine description specific power_save 3773 function to NULL. On Idle the CPU just reduces 3774 execution priority. 3775 3776 ppc_strict_facility_enable 3777 [PPC] This option catches any kernel floating point, 3778 Altivec, VSX and SPE outside of regions specifically 3779 allowed (eg kernel_enable_fpu()/kernel_disable_fpu()). 3780 There is some performance impact when enabling this. 3781 3782 ppc_tm= [PPC] 3783 Format: {"off"} 3784 Disable Hardware Transactional Memory 3785 3786 print-fatal-signals= 3787 [KNL] debug: print fatal signals 3788 3789 If enabled, warn about various signal handling 3790 related application anomalies: too many signals, 3791 too many POSIX.1 timers, fatal signals causing a 3792 coredump - etc. 3793 3794 If you hit the warning due to signal overflow, 3795 you might want to try "ulimit -i unlimited". 3796 3797 default: off. 3798 3799 printk.always_kmsg_dump= 3800 Trigger kmsg_dump for cases other than kernel oops or 3801 panics 3802 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 3803 default: disabled 3804 3805 printk.devkmsg={on,off,ratelimit} 3806 Control writing to /dev/kmsg. 3807 on - unlimited logging to /dev/kmsg from userspace 3808 off - logging to /dev/kmsg disabled 3809 ratelimit - ratelimit the logging 3810 Default: ratelimit 3811 3812 printk.time= Show timing data prefixed to each printk message line 3813 Format: <bool> (1/Y/y=enable, 0/N/n=disable) 3814 3815 processor.max_cstate= [HW,ACPI] 3816 Limit processor to maximum C-state 3817 max_cstate=9 overrides any DMI blacklist limit. 3818 3819 processor.nocst [HW,ACPI] 3820 Ignore the _CST method to determine C-states, 3821 instead using the legacy FADT method 3822 3823 profile= [KNL] Enable kernel profiling via /proc/profile 3824 Format: [<profiletype>,]<number> 3825 Param: <profiletype>: "schedule", "sleep", or "kvm" 3826 [defaults to kernel profiling] 3827 Param: "schedule" - profile schedule points. 3828 Param: "sleep" - profile D-state sleeping (millisecs). 3829 Requires CONFIG_SCHEDSTATS 3830 Param: "kvm" - profile VM exits. 3831 Param: <number> - step/bucket size as a power of 2 for 3832 statistical time based profiling. 3833 3834 prompt_ramdisk= [RAM] List of RAM disks to prompt for floppy disk 3835 before loading. 3836 See Documentation/admin-guide/blockdev/ramdisk.rst. 3837 3838 prot_virt= [S390] enable hosting protected virtual machines 3839 isolated from the hypervisor (if hardware supports 3840 that). 3841 Format: <bool> 3842 3843 psi= [KNL] Enable or disable pressure stall information 3844 tracking. 3845 Format: <bool> 3846 3847 psmouse.proto= [HW,MOUSE] Highest PS2 mouse protocol extension to 3848 probe for; one of (bare|imps|exps|lifebook|any). 3849 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports 3850 per second. 3851 psmouse.resetafter= [HW,MOUSE] 3852 Try to reset the device after so many bad packets 3853 (0 = never). 3854 psmouse.resolution= 3855 [HW,MOUSE] Set desired mouse resolution, in dpi. 3856 psmouse.smartscroll= 3857 [HW,MOUSE] Controls Logitech smartscroll autorepeat. 3858 0 = disabled, 1 = enabled (default). 3859 3860 pstore.backend= Specify the name of the pstore backend to use 3861 3862 pt. [PARIDE] 3863 See Documentation/admin-guide/blockdev/paride.rst. 3864 3865 pti= [X86_64] Control Page Table Isolation of user and 3866 kernel address spaces. Disabling this feature 3867 removes hardening, but improves performance of 3868 system calls and interrupts. 3869 3870 on - unconditionally enable 3871 off - unconditionally disable 3872 auto - kernel detects whether your CPU model is 3873 vulnerable to issues that PTI mitigates 3874 3875 Not specifying this option is equivalent to pti=auto. 3876 3877 nopti [X86_64] 3878 Equivalent to pti=off 3879 3880 pty.legacy_count= 3881 [KNL] Number of legacy pty's. Overwrites compiled-in 3882 default number. 3883 3884 quiet [KNL] Disable most log messages 3885 3886 r128= [HW,DRM] 3887 3888 raid= [HW,RAID] 3889 See Documentation/admin-guide/md.rst. 3890 3891 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes 3892 See Documentation/admin-guide/blockdev/ramdisk.rst. 3893 3894 random.trust_cpu={on,off} 3895 [KNL] Enable or disable trusting the use of the 3896 CPU's random number generator (if available) to 3897 fully seed the kernel's CRNG. Default is controlled 3898 by CONFIG_RANDOM_TRUST_CPU. 3899 3900 ras=option[,option,...] [KNL] RAS-specific options 3901 3902 cec_disable [X86] 3903 Disable the Correctable Errors Collector, 3904 see CONFIG_RAS_CEC help text. 3905 3906 rcu_nocbs= [KNL] 3907 The argument is a cpu list, as described above, 3908 except that the string "all" can be used to 3909 specify every CPU on the system. 3910 3911 In kernels built with CONFIG_RCU_NOCB_CPU=y, set 3912 the specified list of CPUs to be no-callback CPUs. 3913 Invocation of these CPUs' RCU callbacks will be 3914 offloaded to "rcuox/N" kthreads created for that 3915 purpose, where "x" is "p" for RCU-preempt, and 3916 "s" for RCU-sched, and "N" is the CPU number. 3917 This reduces OS jitter on the offloaded CPUs, 3918 which can be useful for HPC and real-time 3919 workloads. It can also improve energy efficiency 3920 for asymmetric multiprocessors. 3921 3922 rcu_nocb_poll [KNL] 3923 Rather than requiring that offloaded CPUs 3924 (specified by rcu_nocbs= above) explicitly 3925 awaken the corresponding "rcuoN" kthreads, 3926 make these kthreads poll for callbacks. 3927 This improves the real-time response for the 3928 offloaded CPUs by relieving them of the need to 3929 wake up the corresponding kthread, but degrades 3930 energy efficiency by requiring that the kthreads 3931 periodically wake up to do the polling. 3932 3933 rcutree.blimit= [KNL] 3934 Set maximum number of finished RCU callbacks to 3935 process in one batch. 3936 3937 rcutree.dump_tree= [KNL] 3938 Dump the structure of the rcu_node combining tree 3939 out at early boot. This is used for diagnostic 3940 purposes, to verify correct tree setup. 3941 3942 rcutree.gp_cleanup_delay= [KNL] 3943 Set the number of jiffies to delay each step of 3944 RCU grace-period cleanup. 3945 3946 rcutree.gp_init_delay= [KNL] 3947 Set the number of jiffies to delay each step of 3948 RCU grace-period initialization. 3949 3950 rcutree.gp_preinit_delay= [KNL] 3951 Set the number of jiffies to delay each step of 3952 RCU grace-period pre-initialization, that is, 3953 the propagation of recent CPU-hotplug changes up 3954 the rcu_node combining tree. 3955 3956 rcutree.use_softirq= [KNL] 3957 If set to zero, move all RCU_SOFTIRQ processing to 3958 per-CPU rcuc kthreads. Defaults to a non-zero 3959 value, meaning that RCU_SOFTIRQ is used by default. 3960 Specify rcutree.use_softirq=0 to use rcuc kthreads. 3961 3962 rcutree.rcu_fanout_exact= [KNL] 3963 Disable autobalancing of the rcu_node combining 3964 tree. This is used by rcutorture, and might 3965 possibly be useful for architectures having high 3966 cache-to-cache transfer latencies. 3967 3968 rcutree.rcu_fanout_leaf= [KNL] 3969 Change the number of CPUs assigned to each 3970 leaf rcu_node structure. Useful for very 3971 large systems, which will choose the value 64, 3972 and for NUMA systems with large remote-access 3973 latencies, which will choose a value aligned 3974 with the appropriate hardware boundaries. 3975 3976 rcutree.jiffies_till_first_fqs= [KNL] 3977 Set delay from grace-period initialization to 3978 first attempt to force quiescent states. 3979 Units are jiffies, minimum value is zero, 3980 and maximum value is HZ. 3981 3982 rcutree.jiffies_till_next_fqs= [KNL] 3983 Set delay between subsequent attempts to force 3984 quiescent states. Units are jiffies, minimum 3985 value is one, and maximum value is HZ. 3986 3987 rcutree.jiffies_till_sched_qs= [KNL] 3988 Set required age in jiffies for a 3989 given grace period before RCU starts 3990 soliciting quiescent-state help from 3991 rcu_note_context_switch() and cond_resched(). 3992 If not specified, the kernel will calculate 3993 a value based on the most recent settings 3994 of rcutree.jiffies_till_first_fqs 3995 and rcutree.jiffies_till_next_fqs. 3996 This calculated value may be viewed in 3997 rcutree.jiffies_to_sched_qs. Any attempt to set 3998 rcutree.jiffies_to_sched_qs will be cheerfully 3999 overwritten. 4000 4001 rcutree.kthread_prio= [KNL,BOOT] 4002 Set the SCHED_FIFO priority of the RCU per-CPU 4003 kthreads (rcuc/N). This value is also used for 4004 the priority of the RCU boost threads (rcub/N) 4005 and for the RCU grace-period kthreads (rcu_bh, 4006 rcu_preempt, and rcu_sched). If RCU_BOOST is 4007 set, valid values are 1-99 and the default is 1 4008 (the least-favored priority). Otherwise, when 4009 RCU_BOOST is not set, valid values are 0-99 and 4010 the default is zero (non-realtime operation). 4011 4012 rcutree.rcu_nocb_gp_stride= [KNL] 4013 Set the number of NOCB callback kthreads in 4014 each group, which defaults to the square root 4015 of the number of CPUs. Larger numbers reduce 4016 the wakeup overhead on the global grace-period 4017 kthread, but increases that same overhead on 4018 each group's NOCB grace-period kthread. 4019 4020 rcutree.qhimark= [KNL] 4021 Set threshold of queued RCU callbacks beyond which 4022 batch limiting is disabled. 4023 4024 rcutree.qlowmark= [KNL] 4025 Set threshold of queued RCU callbacks below which 4026 batch limiting is re-enabled. 4027 4028 rcutree.qovld= [KNL] 4029 Set threshold of queued RCU callbacks beyond which 4030 RCU's force-quiescent-state scan will aggressively 4031 enlist help from cond_resched() and sched IPIs to 4032 help CPUs more quickly reach quiescent states. 4033 Set to less than zero to make this be set based 4034 on rcutree.qhimark at boot time and to zero to 4035 disable more aggressive help enlistment. 4036 4037 rcutree.rcu_idle_gp_delay= [KNL] 4038 Set wakeup interval for idle CPUs that have 4039 RCU callbacks (RCU_FAST_NO_HZ=y). 4040 4041 rcutree.rcu_idle_lazy_gp_delay= [KNL] 4042 Set wakeup interval for idle CPUs that have 4043 only "lazy" RCU callbacks (RCU_FAST_NO_HZ=y). 4044 Lazy RCU callbacks are those which RCU can 4045 prove do nothing more than free memory. 4046 4047 rcutree.rcu_kick_kthreads= [KNL] 4048 Cause the grace-period kthread to get an extra 4049 wake_up() if it sleeps three times longer than 4050 it should at force-quiescent-state time. 4051 This wake_up() will be accompanied by a 4052 WARN_ONCE() splat and an ftrace_dump(). 4053 4054 rcutree.sysrq_rcu= [KNL] 4055 Commandeer a sysrq key to dump out Tree RCU's 4056 rcu_node tree with an eye towards determining 4057 why a new grace period has not yet started. 4058 4059 rcuperf.gp_async= [KNL] 4060 Measure performance of asynchronous 4061 grace-period primitives such as call_rcu(). 4062 4063 rcuperf.gp_async_max= [KNL] 4064 Specify the maximum number of outstanding 4065 callbacks per writer thread. When a writer 4066 thread exceeds this limit, it invokes the 4067 corresponding flavor of rcu_barrier() to allow 4068 previously posted callbacks to drain. 4069 4070 rcuperf.gp_exp= [KNL] 4071 Measure performance of expedited synchronous 4072 grace-period primitives. 4073 4074 rcuperf.holdoff= [KNL] 4075 Set test-start holdoff period. The purpose of 4076 this parameter is to delay the start of the 4077 test until boot completes in order to avoid 4078 interference. 4079 4080 rcuperf.kfree_rcu_test= [KNL] 4081 Set to measure performance of kfree_rcu() flooding. 4082 4083 rcuperf.kfree_nthreads= [KNL] 4084 The number of threads running loops of kfree_rcu(). 4085 4086 rcuperf.kfree_alloc_num= [KNL] 4087 Number of allocations and frees done in an iteration. 4088 4089 rcuperf.kfree_loops= [KNL] 4090 Number of loops doing rcuperf.kfree_alloc_num number 4091 of allocations and frees. 4092 4093 rcuperf.nreaders= [KNL] 4094 Set number of RCU readers. The value -1 selects 4095 N, where N is the number of CPUs. A value 4096 "n" less than -1 selects N-n+1, where N is again 4097 the number of CPUs. For example, -2 selects N 4098 (the number of CPUs), -3 selects N+1, and so on. 4099 A value of "n" less than or equal to -N selects 4100 a single reader. 4101 4102 rcuperf.nwriters= [KNL] 4103 Set number of RCU writers. The values operate 4104 the same as for rcuperf.nreaders. 4105 N, where N is the number of CPUs 4106 4107 rcuperf.perf_type= [KNL] 4108 Specify the RCU implementation to test. 4109 4110 rcuperf.shutdown= [KNL] 4111 Shut the system down after performance tests 4112 complete. This is useful for hands-off automated 4113 testing. 4114 4115 rcuperf.verbose= [KNL] 4116 Enable additional printk() statements. 4117 4118 rcuperf.writer_holdoff= [KNL] 4119 Write-side holdoff between grace periods, 4120 in microseconds. The default of zero says 4121 no holdoff. 4122 4123 rcutorture.fqs_duration= [KNL] 4124 Set duration of force_quiescent_state bursts 4125 in microseconds. 4126 4127 rcutorture.fqs_holdoff= [KNL] 4128 Set holdoff time within force_quiescent_state bursts 4129 in microseconds. 4130 4131 rcutorture.fqs_stutter= [KNL] 4132 Set wait time between force_quiescent_state bursts 4133 in seconds. 4134 4135 rcutorture.fwd_progress= [KNL] 4136 Enable RCU grace-period forward-progress testing 4137 for the types of RCU supporting this notion. 4138 4139 rcutorture.fwd_progress_div= [KNL] 4140 Specify the fraction of a CPU-stall-warning 4141 period to do tight-loop forward-progress testing. 4142 4143 rcutorture.fwd_progress_holdoff= [KNL] 4144 Number of seconds to wait between successive 4145 forward-progress tests. 4146 4147 rcutorture.fwd_progress_need_resched= [KNL] 4148 Enclose cond_resched() calls within checks for 4149 need_resched() during tight-loop forward-progress 4150 testing. 4151 4152 rcutorture.gp_cond= [KNL] 4153 Use conditional/asynchronous update-side 4154 primitives, if available. 4155 4156 rcutorture.gp_exp= [KNL] 4157 Use expedited update-side primitives, if available. 4158 4159 rcutorture.gp_normal= [KNL] 4160 Use normal (non-expedited) asynchronous 4161 update-side primitives, if available. 4162 4163 rcutorture.gp_sync= [KNL] 4164 Use normal (non-expedited) synchronous 4165 update-side primitives, if available. If all 4166 of rcutorture.gp_cond=, rcutorture.gp_exp=, 4167 rcutorture.gp_normal=, and rcutorture.gp_sync= 4168 are zero, rcutorture acts as if is interpreted 4169 they are all non-zero. 4170 4171 rcutorture.n_barrier_cbs= [KNL] 4172 Set callbacks/threads for rcu_barrier() testing. 4173 4174 rcutorture.nfakewriters= [KNL] 4175 Set number of concurrent RCU writers. These just 4176 stress RCU, they don't participate in the actual 4177 test, hence the "fake". 4178 4179 rcutorture.nreaders= [KNL] 4180 Set number of RCU readers. The value -1 selects 4181 N-1, where N is the number of CPUs. A value 4182 "n" less than -1 selects N-n-2, where N is again 4183 the number of CPUs. For example, -2 selects N 4184 (the number of CPUs), -3 selects N+1, and so on. 4185 4186 rcutorture.object_debug= [KNL] 4187 Enable debug-object double-call_rcu() testing. 4188 4189 rcutorture.onoff_holdoff= [KNL] 4190 Set time (s) after boot for CPU-hotplug testing. 4191 4192 rcutorture.onoff_interval= [KNL] 4193 Set time (jiffies) between CPU-hotplug operations, 4194 or zero to disable CPU-hotplug testing. 4195 4196 rcutorture.shuffle_interval= [KNL] 4197 Set task-shuffle interval (s). Shuffling tasks 4198 allows some CPUs to go into dyntick-idle mode 4199 during the rcutorture test. 4200 4201 rcutorture.shutdown_secs= [KNL] 4202 Set time (s) after boot system shutdown. This 4203 is useful for hands-off automated testing. 4204 4205 rcutorture.stall_cpu= [KNL] 4206 Duration of CPU stall (s) to test RCU CPU stall 4207 warnings, zero to disable. 4208 4209 rcutorture.stall_cpu_holdoff= [KNL] 4210 Time to wait (s) after boot before inducing stall. 4211 4212 rcutorture.stall_cpu_irqsoff= [KNL] 4213 Disable interrupts while stalling if set. 4214 4215 rcutorture.stat_interval= [KNL] 4216 Time (s) between statistics printk()s. 4217 4218 rcutorture.stutter= [KNL] 4219 Time (s) to stutter testing, for example, specifying 4220 five seconds causes the test to run for five seconds, 4221 wait for five seconds, and so on. This tests RCU's 4222 ability to transition abruptly to and from idle. 4223 4224 rcutorture.test_boost= [KNL] 4225 Test RCU priority boosting? 0=no, 1=maybe, 2=yes. 4226 "Maybe" means test if the RCU implementation 4227 under test support RCU priority boosting. 4228 4229 rcutorture.test_boost_duration= [KNL] 4230 Duration (s) of each individual boost test. 4231 4232 rcutorture.test_boost_interval= [KNL] 4233 Interval (s) between each boost test. 4234 4235 rcutorture.test_no_idle_hz= [KNL] 4236 Test RCU's dyntick-idle handling. See also the 4237 rcutorture.shuffle_interval parameter. 4238 4239 rcutorture.torture_type= [KNL] 4240 Specify the RCU implementation to test. 4241 4242 rcutorture.verbose= [KNL] 4243 Enable additional printk() statements. 4244 4245 rcupdate.rcu_cpu_stall_ftrace_dump= [KNL] 4246 Dump ftrace buffer after reporting RCU CPU 4247 stall warning. 4248 4249 rcupdate.rcu_cpu_stall_suppress= [KNL] 4250 Suppress RCU CPU stall warning messages. 4251 4252 rcupdate.rcu_cpu_stall_suppress_at_boot= [KNL] 4253 Suppress RCU CPU stall warning messages and 4254 rcutorture writer stall warnings that occur 4255 during early boot, that is, during the time 4256 before the init task is spawned. 4257 4258 rcupdate.rcu_cpu_stall_timeout= [KNL] 4259 Set timeout for RCU CPU stall warning messages. 4260 4261 rcupdate.rcu_expedited= [KNL] 4262 Use expedited grace-period primitives, for 4263 example, synchronize_rcu_expedited() instead 4264 of synchronize_rcu(). This reduces latency, 4265 but can increase CPU utilization, degrade 4266 real-time latency, and degrade energy efficiency. 4267 No effect on CONFIG_TINY_RCU kernels. 4268 4269 rcupdate.rcu_normal= [KNL] 4270 Use only normal grace-period primitives, 4271 for example, synchronize_rcu() instead of 4272 synchronize_rcu_expedited(). This improves 4273 real-time latency, CPU utilization, and 4274 energy efficiency, but can expose users to 4275 increased grace-period latency. This parameter 4276 overrides rcupdate.rcu_expedited. No effect on 4277 CONFIG_TINY_RCU kernels. 4278 4279 rcupdate.rcu_normal_after_boot= [KNL] 4280 Once boot has completed (that is, after 4281 rcu_end_inkernel_boot() has been invoked), use 4282 only normal grace-period primitives. No effect 4283 on CONFIG_TINY_RCU kernels. 4284 4285 rcupdate.rcu_task_stall_timeout= [KNL] 4286 Set timeout in jiffies for RCU task stall warning 4287 messages. Disable with a value less than or equal 4288 to zero. 4289 4290 rcupdate.rcu_self_test= [KNL] 4291 Run the RCU early boot self tests 4292 4293 rdinit= [KNL] 4294 Format: <full_path> 4295 Run specified binary instead of /init from the ramdisk, 4296 used for early userspace startup. See initrd. 4297 4298 rdrand= [X86] 4299 force - Override the decision by the kernel to hide the 4300 advertisement of RDRAND support (this affects 4301 certain AMD processors because of buggy BIOS 4302 support, specifically around the suspend/resume 4303 path). 4304 4305 rdt= [HW,X86,RDT] 4306 Turn on/off individual RDT features. List is: 4307 cmt, mbmtotal, mbmlocal, l3cat, l3cdp, l2cat, l2cdp, 4308 mba. 4309 E.g. to turn on cmt and turn off mba use: 4310 rdt=cmt,!mba 4311 4312 reboot= [KNL] 4313 Format (x86 or x86_64): 4314 [w[arm] | c[old] | h[ard] | s[oft] | g[pio]] \ 4315 [[,]s[mp]#### \ 4316 [[,]b[ios] | a[cpi] | k[bd] | t[riple] | e[fi] | p[ci]] \ 4317 [[,]f[orce] 4318 Where reboot_mode is one of warm (soft) or cold (hard) or gpio 4319 (prefix with 'panic_' to set mode for panic 4320 reboot only), 4321 reboot_type is one of bios, acpi, kbd, triple, efi, or pci, 4322 reboot_force is either force or not specified, 4323 reboot_cpu is s[mp]#### with #### being the processor 4324 to be used for rebooting. 4325 4326 relax_domain_level= 4327 [KNL, SMP] Set scheduler's default relax_domain_level. 4328 See Documentation/admin-guide/cgroup-v1/cpusets.rst. 4329 4330 reserve= [KNL,BUGS] Force kernel to ignore I/O ports or memory 4331 Format: <base1>,<size1>[,<base2>,<size2>,...] 4332 Reserve I/O ports or memory so the kernel won't use 4333 them. If <base> is less than 0x10000, the region 4334 is assumed to be I/O ports; otherwise it is memory. 4335 4336 reservetop= [X86-32] 4337 Format: nn[KMG] 4338 Reserves a hole at the top of the kernel virtual 4339 address space. 4340 4341 reservelow= [X86] 4342 Format: nn[K] 4343 Set the amount of memory to reserve for BIOS at 4344 the bottom of the address space. 4345 4346 reset_devices [KNL] Force drivers to reset the underlying device 4347 during initialization. 4348 4349 resume= [SWSUSP] 4350 Specify the partition device for software suspend 4351 Format: 4352 {/dev/<dev> | PARTUUID=<uuid> | <int>:<int> | <hex>} 4353 4354 resume_offset= [SWSUSP] 4355 Specify the offset from the beginning of the partition 4356 given by "resume=" at which the swap header is located, 4357 in <PAGE_SIZE> units (needed only for swap files). 4358 See Documentation/power/swsusp-and-swap-files.rst 4359 4360 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to 4361 read the resume files 4362 4363 resumewait [HIBERNATION] Wait (indefinitely) for resume device to show up. 4364 Useful for devices that are detected asynchronously 4365 (e.g. USB and MMC devices). 4366 4367 hibernate= [HIBERNATION] 4368 noresume Don't check if there's a hibernation image 4369 present during boot. 4370 nocompress Don't compress/decompress hibernation images. 4371 no Disable hibernation and resume. 4372 protect_image Turn on image protection during restoration 4373 (that will set all pages holding image data 4374 during restoration read-only). 4375 4376 retain_initrd [RAM] Keep initrd memory after extraction 4377 4378 rfkill.default_state= 4379 0 "airplane mode". All wifi, bluetooth, wimax, gps, fm, 4380 etc. communication is blocked by default. 4381 1 Unblocked. 4382 4383 rfkill.master_switch_mode= 4384 0 The "airplane mode" button does nothing. 4385 1 The "airplane mode" button toggles between everything 4386 blocked and the previous configuration. 4387 2 The "airplane mode" button toggles between everything 4388 blocked and everything unblocked. 4389 4390 rhash_entries= [KNL,NET] 4391 Set number of hash buckets for route cache 4392 4393 ring3mwait=disable 4394 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported 4395 CPUs. 4396 4397 ro [KNL] Mount root device read-only on boot 4398 4399 rodata= [KNL] 4400 on Mark read-only kernel memory as read-only (default). 4401 off Leave read-only kernel memory writable for debugging. 4402 4403 rockchip.usb_uart 4404 Enable the uart passthrough on the designated usb port 4405 on Rockchip SoCs. When active, the signals of the 4406 debug-uart get routed to the D+ and D- pins of the usb 4407 port and the regular usb controller gets disabled. 4408 4409 root= [KNL] Root filesystem 4410 See name_to_dev_t comment in init/do_mounts.c. 4411 4412 rootdelay= [KNL] Delay (in seconds) to pause before attempting to 4413 mount the root filesystem 4414 4415 rootflags= [KNL] Set root filesystem mount option string 4416 4417 rootfstype= [KNL] Set root filesystem type 4418 4419 rootwait [KNL] Wait (indefinitely) for root device to show up. 4420 Useful for devices that are detected asynchronously 4421 (e.g. USB and MMC devices). 4422 4423 rproc_mem=nn[KMG][@address] 4424 [KNL,ARM,CMA] Remoteproc physical memory block. 4425 Memory area to be used by remote processor image, 4426 managed by CMA. 4427 4428 rw [KNL] Mount root device read-write on boot 4429 4430 S [KNL] Run init in single mode 4431 4432 s390_iommu= [HW,S390] 4433 Set s390 IOTLB flushing mode 4434 strict 4435 With strict flushing every unmap operation will result in 4436 an IOTLB flush. Default is lazy flushing before reuse, 4437 which is faster. 4438 4439 sa1100ir [NET] 4440 See drivers/net/irda/sa1100_ir.c. 4441 4442 sbni= [NET] Granch SBNI12 leased line adapter 4443 4444 sched_debug [KNL] Enables verbose scheduler debug messages. 4445 4446 schedstats= [KNL,X86] Enable or disable scheduled statistics. 4447 Allowed values are enable and disable. This feature 4448 incurs a small amount of overhead in the scheduler 4449 but is useful for debugging and performance tuning. 4450 4451 sched_thermal_decay_shift= 4452 [KNL, SMP] Set a decay shift for scheduler thermal 4453 pressure signal. Thermal pressure signal follows the 4454 default decay period of other scheduler pelt 4455 signals(usually 32 ms but configurable). Setting 4456 sched_thermal_decay_shift will left shift the decay 4457 period for the thermal pressure signal by the shift 4458 value. 4459 i.e. with the default pelt decay period of 32 ms 4460 sched_thermal_decay_shift thermal pressure decay pr 4461 1 64 ms 4462 2 128 ms 4463 and so on. 4464 Format: integer between 0 and 10 4465 Default is 0. 4466 4467 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate 4468 xtime_lock contention on larger systems, and/or RCU lock 4469 contention on all systems with CONFIG_MAXSMP set. 4470 Format: { "0" | "1" } 4471 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1" 4472 1 -- enable. 4473 Note: increases power consumption, thus should only be 4474 enabled if running jitter sensitive (HPC/RT) workloads. 4475 4476 security= [SECURITY] Choose a legacy "major" security module to 4477 enable at boot. This has been deprecated by the 4478 "lsm=" parameter. 4479 4480 selinux= [SELINUX] Disable or enable SELinux at boot time. 4481 Format: { "0" | "1" } 4482 See security/selinux/Kconfig help text. 4483 0 -- disable. 4484 1 -- enable. 4485 Default value is 1. 4486 4487 apparmor= [APPARMOR] Disable or enable AppArmor at boot time 4488 Format: { "0" | "1" } 4489 See security/apparmor/Kconfig help text 4490 0 -- disable. 4491 1 -- enable. 4492 Default value is set via kernel config option. 4493 4494 serialnumber [BUGS=X86-32] 4495 4496 shapers= [NET] 4497 Maximal number of shapers. 4498 4499 simeth= [IA-64] 4500 simscsi= 4501 4502 slram= [HW,MTD] 4503 4504 slab_nomerge [MM] 4505 Disable merging of slabs with similar size. May be 4506 necessary if there is some reason to distinguish 4507 allocs to different slabs, especially in hardened 4508 environments where the risk of heap overflows and 4509 layout control by attackers can usually be 4510 frustrated by disabling merging. This will reduce 4511 most of the exposure of a heap attack to a single 4512 cache (risks via metadata attacks are mostly 4513 unchanged). Debug options disable merging on their 4514 own. 4515 For more information see Documentation/vm/slub.rst. 4516 4517 slab_max_order= [MM, SLAB] 4518 Determines the maximum allowed order for slabs. 4519 A high setting may cause OOMs due to memory 4520 fragmentation. Defaults to 1 for systems with 4521 more than 32MB of RAM, 0 otherwise. 4522 4523 slub_debug[=options[,slabs]] [MM, SLUB] 4524 Enabling slub_debug allows one to determine the 4525 culprit if slab objects become corrupted. Enabling 4526 slub_debug can create guard zones around objects and 4527 may poison objects when not in use. Also tracks the 4528 last alloc / free. For more information see 4529 Documentation/vm/slub.rst. 4530 4531 slub_memcg_sysfs= [MM, SLUB] 4532 Determines whether to enable sysfs directories for 4533 memory cgroup sub-caches. 1 to enable, 0 to disable. 4534 The default is determined by CONFIG_SLUB_MEMCG_SYSFS_ON. 4535 Enabling this can lead to a very high number of debug 4536 directories and files being created under 4537 /sys/kernel/slub. 4538 4539 slub_max_order= [MM, SLUB] 4540 Determines the maximum allowed order for slabs. 4541 A high setting may cause OOMs due to memory 4542 fragmentation. For more information see 4543 Documentation/vm/slub.rst. 4544 4545 slub_min_objects= [MM, SLUB] 4546 The minimum number of objects per slab. SLUB will 4547 increase the slab order up to slub_max_order to 4548 generate a sufficiently large slab able to contain 4549 the number of objects indicated. The higher the number 4550 of objects the smaller the overhead of tracking slabs 4551 and the less frequently locks need to be acquired. 4552 For more information see Documentation/vm/slub.rst. 4553 4554 slub_min_order= [MM, SLUB] 4555 Determines the minimum page order for slabs. Must be 4556 lower than slub_max_order. 4557 For more information see Documentation/vm/slub.rst. 4558 4559 slub_nomerge [MM, SLUB] 4560 Same with slab_nomerge. This is supported for legacy. 4561 See slab_nomerge for more information. 4562 4563 smart2= [HW] 4564 Format: <io1>[,<io2>[,...,<io8>]] 4565 4566 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices 4567 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port 4568 smsc-ircc2.ircc_sir= [HW] SIR base I/O port 4569 smsc-ircc2.ircc_fir= [HW] FIR base I/O port 4570 smsc-ircc2.ircc_irq= [HW] IRQ line 4571 smsc-ircc2.ircc_dma= [HW] DMA channel 4572 smsc-ircc2.ircc_transceiver= [HW] Transceiver type: 4573 0: Toshiba Satellite 1800 (GP data pin select) 4574 1: Fast pin select (default) 4575 2: ATC IRMode 4576 4577 smt [KNL,S390] Set the maximum number of threads (logical 4578 CPUs) to use per physical CPU on systems capable of 4579 symmetric multithreading (SMT). Will be capped to the 4580 actual hardware limit. 4581 Format: <integer> 4582 Default: -1 (no limit) 4583 4584 softlockup_panic= 4585 [KNL] Should the soft-lockup detector generate panics. 4586 Format: <integer> 4587 4588 A nonzero value instructs the soft-lockup detector 4589 to panic the machine when a soft-lockup occurs. It is 4590 also controlled by the kernel.softlockup_panic sysctl 4591 and CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC, which is the 4592 respective build-time switch to that functionality. 4593 4594 softlockup_all_cpu_backtrace= 4595 [KNL] Should the soft-lockup detector generate 4596 backtraces on all cpus. 4597 Format: <integer> 4598 4599 sonypi.*= [HW] Sony Programmable I/O Control Device driver 4600 See Documentation/admin-guide/laptops/sonypi.rst 4601 4602 spectre_v2= [X86] Control mitigation of Spectre variant 2 4603 (indirect branch speculation) vulnerability. 4604 The default operation protects the kernel from 4605 user space attacks. 4606 4607 on - unconditionally enable, implies 4608 spectre_v2_user=on 4609 off - unconditionally disable, implies 4610 spectre_v2_user=off 4611 auto - kernel detects whether your CPU model is 4612 vulnerable 4613 4614 Selecting 'on' will, and 'auto' may, choose a 4615 mitigation method at run time according to the 4616 CPU, the available microcode, the setting of the 4617 CONFIG_RETPOLINE configuration option, and the 4618 compiler with which the kernel was built. 4619 4620 Selecting 'on' will also enable the mitigation 4621 against user space to user space task attacks. 4622 4623 Selecting 'off' will disable both the kernel and 4624 the user space protections. 4625 4626 Specific mitigations can also be selected manually: 4627 4628 retpoline - replace indirect branches 4629 retpoline,generic - google's original retpoline 4630 retpoline,amd - AMD-specific minimal thunk 4631 4632 Not specifying this option is equivalent to 4633 spectre_v2=auto. 4634 4635 spectre_v2_user= 4636 [X86] Control mitigation of Spectre variant 2 4637 (indirect branch speculation) vulnerability between 4638 user space tasks 4639 4640 on - Unconditionally enable mitigations. Is 4641 enforced by spectre_v2=on 4642 4643 off - Unconditionally disable mitigations. Is 4644 enforced by spectre_v2=off 4645 4646 prctl - Indirect branch speculation is enabled, 4647 but mitigation can be enabled via prctl 4648 per thread. The mitigation control state 4649 is inherited on fork. 4650 4651 prctl,ibpb 4652 - Like "prctl" above, but only STIBP is 4653 controlled per thread. IBPB is issued 4654 always when switching between different user 4655 space processes. 4656 4657 seccomp 4658 - Same as "prctl" above, but all seccomp 4659 threads will enable the mitigation unless 4660 they explicitly opt out. 4661 4662 seccomp,ibpb 4663 - Like "seccomp" above, but only STIBP is 4664 controlled per thread. IBPB is issued 4665 always when switching between different 4666 user space processes. 4667 4668 auto - Kernel selects the mitigation depending on 4669 the available CPU features and vulnerability. 4670 4671 Default mitigation: 4672 If CONFIG_SECCOMP=y then "seccomp", otherwise "prctl" 4673 4674 Not specifying this option is equivalent to 4675 spectre_v2_user=auto. 4676 4677 spec_store_bypass_disable= 4678 [HW] Control Speculative Store Bypass (SSB) Disable mitigation 4679 (Speculative Store Bypass vulnerability) 4680 4681 Certain CPUs are vulnerable to an exploit against a 4682 a common industry wide performance optimization known 4683 as "Speculative Store Bypass" in which recent stores 4684 to the same memory location may not be observed by 4685 later loads during speculative execution. The idea 4686 is that such stores are unlikely and that they can 4687 be detected prior to instruction retirement at the 4688 end of a particular speculation execution window. 4689 4690 In vulnerable processors, the speculatively forwarded 4691 store can be used in a cache side channel attack, for 4692 example to read memory to which the attacker does not 4693 directly have access (e.g. inside sandboxed code). 4694 4695 This parameter controls whether the Speculative Store 4696 Bypass optimization is used. 4697 4698 On x86 the options are: 4699 4700 on - Unconditionally disable Speculative Store Bypass 4701 off - Unconditionally enable Speculative Store Bypass 4702 auto - Kernel detects whether the CPU model contains an 4703 implementation of Speculative Store Bypass and 4704 picks the most appropriate mitigation. If the 4705 CPU is not vulnerable, "off" is selected. If the 4706 CPU is vulnerable the default mitigation is 4707 architecture and Kconfig dependent. See below. 4708 prctl - Control Speculative Store Bypass per thread 4709 via prctl. Speculative Store Bypass is enabled 4710 for a process by default. The state of the control 4711 is inherited on fork. 4712 seccomp - Same as "prctl" above, but all seccomp threads 4713 will disable SSB unless they explicitly opt out. 4714 4715 Default mitigations: 4716 X86: If CONFIG_SECCOMP=y "seccomp", otherwise "prctl" 4717 4718 On powerpc the options are: 4719 4720 on,auto - On Power8 and Power9 insert a store-forwarding 4721 barrier on kernel entry and exit. On Power7 4722 perform a software flush on kernel entry and 4723 exit. 4724 off - No action. 4725 4726 Not specifying this option is equivalent to 4727 spec_store_bypass_disable=auto. 4728 4729 spia_io_base= [HW,MTD] 4730 spia_fio_base= 4731 spia_pedr= 4732 spia_peddr= 4733 4734 split_lock_detect= 4735 [X86] Enable split lock detection 4736 4737 When enabled (and if hardware support is present), atomic 4738 instructions that access data across cache line 4739 boundaries will result in an alignment check exception. 4740 4741 off - not enabled 4742 4743 warn - the kernel will emit rate limited warnings 4744 about applications triggering the #AC 4745 exception. This mode is the default on CPUs 4746 that supports split lock detection. 4747 4748 fatal - the kernel will send SIGBUS to applications 4749 that trigger the #AC exception. 4750 4751 If an #AC exception is hit in the kernel or in 4752 firmware (i.e. not while executing in user mode) 4753 the kernel will oops in either "warn" or "fatal" 4754 mode. 4755 4756 srcutree.counter_wrap_check [KNL] 4757 Specifies how frequently to check for 4758 grace-period sequence counter wrap for the 4759 srcu_data structure's ->srcu_gp_seq_needed field. 4760 The greater the number of bits set in this kernel 4761 parameter, the less frequently counter wrap will 4762 be checked for. Note that the bottom two bits 4763 are ignored. 4764 4765 srcutree.exp_holdoff [KNL] 4766 Specifies how many nanoseconds must elapse 4767 since the end of the last SRCU grace period for 4768 a given srcu_struct until the next normal SRCU 4769 grace period will be considered for automatic 4770 expediting. Set to zero to disable automatic 4771 expediting. 4772 4773 ssbd= [ARM64,HW] 4774 Speculative Store Bypass Disable control 4775 4776 On CPUs that are vulnerable to the Speculative 4777 Store Bypass vulnerability and offer a 4778 firmware based mitigation, this parameter 4779 indicates how the mitigation should be used: 4780 4781 force-on: Unconditionally enable mitigation for 4782 for both kernel and userspace 4783 force-off: Unconditionally disable mitigation for 4784 for both kernel and userspace 4785 kernel: Always enable mitigation in the 4786 kernel, and offer a prctl interface 4787 to allow userspace to register its 4788 interest in being mitigated too. 4789 4790 stack_guard_gap= [MM] 4791 override the default stack gap protection. The value 4792 is in page units and it defines how many pages prior 4793 to (for stacks growing down) resp. after (for stacks 4794 growing up) the main stack are reserved for no other 4795 mapping. Default value is 256 pages. 4796 4797 stacktrace [FTRACE] 4798 Enabled the stack tracer on boot up. 4799 4800 stacktrace_filter=[function-list] 4801 [FTRACE] Limit the functions that the stack tracer 4802 will trace at boot up. function-list is a comma separated 4803 list of functions. This list can be changed at run 4804 time by the stack_trace_filter file in the debugfs 4805 tracing directory. Note, this enables stack tracing 4806 and the stacktrace above is not needed. 4807 4808 sti= [PARISC,HW] 4809 Format: <num> 4810 Set the STI (builtin display/keyboard on the HP-PARISC 4811 machines) console (graphic card) which should be used 4812 as the initial boot-console. 4813 See also comment in drivers/video/console/sticore.c. 4814 4815 sti_font= [HW] 4816 See comment in drivers/video/console/sticore.c. 4817 4818 stifb= [HW] 4819 Format: bpp:<bpp1>[:<bpp2>[:<bpp3>...]] 4820 4821 sunrpc.min_resvport= 4822 sunrpc.max_resvport= 4823 [NFS,SUNRPC] 4824 SunRPC servers often require that client requests 4825 originate from a privileged port (i.e. a port in the 4826 range 0 < portnr < 1024). 4827 An administrator who wishes to reserve some of these 4828 ports for other uses may adjust the range that the 4829 kernel's sunrpc client considers to be privileged 4830 using these two parameters to set the minimum and 4831 maximum port values. 4832 4833 sunrpc.svc_rpc_per_connection_limit= 4834 [NFS,SUNRPC] 4835 Limit the number of requests that the server will 4836 process in parallel from a single connection. 4837 The default value is 0 (no limit). 4838 4839 sunrpc.pool_mode= 4840 [NFS] 4841 Control how the NFS server code allocates CPUs to 4842 service thread pools. Depending on how many NICs 4843 you have and where their interrupts are bound, this 4844 option will affect which CPUs will do NFS serving. 4845 Note: this parameter cannot be changed while the 4846 NFS server is running. 4847 4848 auto the server chooses an appropriate mode 4849 automatically using heuristics 4850 global a single global pool contains all CPUs 4851 percpu one pool for each CPU 4852 pernode one pool for each NUMA node (equivalent 4853 to global on non-NUMA machines) 4854 4855 sunrpc.tcp_slot_table_entries= 4856 sunrpc.udp_slot_table_entries= 4857 [NFS,SUNRPC] 4858 Sets the upper limit on the number of simultaneous 4859 RPC calls that can be sent from the client to a 4860 server. Increasing these values may allow you to 4861 improve throughput, but will also increase the 4862 amount of memory reserved for use by the client. 4863 4864 suspend.pm_test_delay= 4865 [SUSPEND] 4866 Sets the number of seconds to remain in a suspend test 4867 mode before resuming the system (see 4868 /sys/power/pm_test). Only available when CONFIG_PM_DEBUG 4869 is set. Default value is 5. 4870 4871 svm= [PPC] 4872 Format: { on | off | y | n | 1 | 0 } 4873 This parameter controls use of the Protected 4874 Execution Facility on pSeries. 4875 4876 swapaccount=[0|1] 4877 [KNL] Enable accounting of swap in memory resource 4878 controller if no parameter or 1 is given or disable 4879 it if 0 is given (See Documentation/admin-guide/cgroup-v1/memory.rst) 4880 4881 swiotlb= [ARM,IA-64,PPC,MIPS,X86] 4882 Format: { <int> | force | noforce } 4883 <int> -- Number of I/O TLB slabs 4884 force -- force using of bounce buffers even if they 4885 wouldn't be automatically used by the kernel 4886 noforce -- Never use bounce buffers (for debugging) 4887 4888 switches= [HW,M68k] 4889 4890 sysfs.deprecated=0|1 [KNL] 4891 Enable/disable old style sysfs layout for old udev 4892 on older distributions. When this option is enabled 4893 very new udev will not work anymore. When this option 4894 is disabled (or CONFIG_SYSFS_DEPRECATED not compiled) 4895 in older udev will not work anymore. 4896 Default depends on CONFIG_SYSFS_DEPRECATED_V2 set in 4897 the kernel configuration. 4898 4899 sysrq_always_enabled 4900 [KNL] 4901 Ignore sysrq setting - this boot parameter will 4902 neutralize any effect of /proc/sys/kernel/sysrq. 4903 Useful for debugging. 4904 4905 tcpmhash_entries= [KNL,NET] 4906 Set the number of tcp_metrics_hash slots. 4907 Default value is 8192 or 16384 depending on total 4908 ram pages. This is used to specify the TCP metrics 4909 cache size. See Documentation/networking/ip-sysctl.txt 4910 "tcp_no_metrics_save" section for more details. 4911 4912 tdfx= [HW,DRM] 4913 4914 test_suspend= [SUSPEND][,N] 4915 Specify "mem" (for Suspend-to-RAM) or "standby" (for 4916 standby suspend) or "freeze" (for suspend type freeze) 4917 as the system sleep state during system startup with 4918 the optional capability to repeat N number of times. 4919 The system is woken from this state using a 4920 wakeup-capable RTC alarm. 4921 4922 thash_entries= [KNL,NET] 4923 Set number of hash buckets for TCP connection 4924 4925 thermal.act= [HW,ACPI] 4926 -1: disable all active trip points in all thermal zones 4927 <degrees C>: override all lowest active trip points 4928 4929 thermal.crt= [HW,ACPI] 4930 -1: disable all critical trip points in all thermal zones 4931 <degrees C>: override all critical trip points 4932 4933 thermal.nocrt= [HW,ACPI] 4934 Set to disable actions on ACPI thermal zone 4935 critical and hot trip points. 4936 4937 thermal.off= [HW,ACPI] 4938 1: disable ACPI thermal control 4939 4940 thermal.psv= [HW,ACPI] 4941 -1: disable all passive trip points 4942 <degrees C>: override all passive trip points to this 4943 value 4944 4945 thermal.tzp= [HW,ACPI] 4946 Specify global default ACPI thermal zone polling rate 4947 <deci-seconds>: poll all this frequency 4948 0: no polling (default) 4949 4950 threadirqs [KNL] 4951 Force threading of all interrupt handlers except those 4952 marked explicitly IRQF_NO_THREAD. 4953 4954 topology= [S390] 4955 Format: {off | on} 4956 Specify if the kernel should make use of the cpu 4957 topology information if the hardware supports this. 4958 The scheduler will make use of this information and 4959 e.g. base its process migration decisions on it. 4960 Default is on. 4961 4962 topology_updates= [KNL, PPC, NUMA] 4963 Format: {off} 4964 Specify if the kernel should ignore (off) 4965 topology updates sent by the hypervisor to this 4966 LPAR. 4967 4968 torture.disable_onoff_at_boot= [KNL] 4969 Prevent the CPU-hotplug component of torturing 4970 until after init has spawned. 4971 4972 tp720= [HW,PS2] 4973 4974 tpm_suspend_pcr=[HW,TPM] 4975 Format: integer pcr id 4976 Specify that at suspend time, the tpm driver 4977 should extend the specified pcr with zeros, 4978 as a workaround for some chips which fail to 4979 flush the last written pcr on TPM_SaveState. 4980 This will guarantee that all the other pcrs 4981 are saved. 4982 4983 trace_buf_size=nn[KMG] 4984 [FTRACE] will set tracing buffer size on each cpu. 4985 4986 trace_event=[event-list] 4987 [FTRACE] Set and start specified trace events in order 4988 to facilitate early boot debugging. The event-list is a 4989 comma separated list of trace events to enable. See 4990 also Documentation/trace/events.rst 4991 4992 trace_options=[option-list] 4993 [FTRACE] Enable or disable tracer options at boot. 4994 The option-list is a comma delimited list of options 4995 that can be enabled or disabled just as if you were 4996 to echo the option name into 4997 4998 /sys/kernel/debug/tracing/trace_options 4999 5000 For example, to enable stacktrace option (to dump the 5001 stack trace of each event), add to the command line: 5002 5003 trace_options=stacktrace 5004 5005 See also Documentation/trace/ftrace.rst "trace options" 5006 section. 5007 5008 tp_printk[FTRACE] 5009 Have the tracepoints sent to printk as well as the 5010 tracing ring buffer. This is useful for early boot up 5011 where the system hangs or reboots and does not give the 5012 option for reading the tracing buffer or performing a 5013 ftrace_dump_on_oops. 5014 5015 To turn off having tracepoints sent to printk, 5016 echo 0 > /proc/sys/kernel/tracepoint_printk 5017 Note, echoing 1 into this file without the 5018 tracepoint_printk kernel cmdline option has no effect. 5019 5020 ** CAUTION ** 5021 5022 Having tracepoints sent to printk() and activating high 5023 frequency tracepoints such as irq or sched, can cause 5024 the system to live lock. 5025 5026 traceoff_on_warning 5027 [FTRACE] enable this option to disable tracing when a 5028 warning is hit. This turns off "tracing_on". Tracing can 5029 be enabled again by echoing '1' into the "tracing_on" 5030 file located in /sys/kernel/debug/tracing/ 5031 5032 This option is useful, as it disables the trace before 5033 the WARNING dump is called, which prevents the trace to 5034 be filled with content caused by the warning output. 5035 5036 This option can also be set at run time via the sysctl 5037 option: kernel/traceoff_on_warning 5038 5039 transparent_hugepage= 5040 [KNL] 5041 Format: [always|madvise|never] 5042 Can be used to control the default behavior of the system 5043 with respect to transparent hugepages. 5044 See Documentation/admin-guide/mm/transhuge.rst 5045 for more details. 5046 5047 tsc= Disable clocksource stability checks for TSC. 5048 Format: <string> 5049 [x86] reliable: mark tsc clocksource as reliable, this 5050 disables clocksource verification at runtime, as well 5051 as the stability checks done at bootup. Used to enable 5052 high-resolution timer mode on older hardware, and in 5053 virtualized environment. 5054 [x86] noirqtime: Do not use TSC to do irq accounting. 5055 Used to run time disable IRQ_TIME_ACCOUNTING on any 5056 platforms where RDTSC is slow and this accounting 5057 can add overhead. 5058 [x86] unstable: mark the TSC clocksource as unstable, this 5059 marks the TSC unconditionally unstable at bootup and 5060 avoids any further wobbles once the TSC watchdog notices. 5061 [x86] nowatchdog: disable clocksource watchdog. Used 5062 in situations with strict latency requirements (where 5063 interruptions from clocksource watchdog are not 5064 acceptable). 5065 5066 tsx= [X86] Control Transactional Synchronization 5067 Extensions (TSX) feature in Intel processors that 5068 support TSX control. 5069 5070 This parameter controls the TSX feature. The options are: 5071 5072 on - Enable TSX on the system. Although there are 5073 mitigations for all known security vulnerabilities, 5074 TSX has been known to be an accelerator for 5075 several previous speculation-related CVEs, and 5076 so there may be unknown security risks associated 5077 with leaving it enabled. 5078 5079 off - Disable TSX on the system. (Note that this 5080 option takes effect only on newer CPUs which are 5081 not vulnerable to MDS, i.e., have 5082 MSR_IA32_ARCH_CAPABILITIES.MDS_NO=1 and which get 5083 the new IA32_TSX_CTRL MSR through a microcode 5084 update. This new MSR allows for the reliable 5085 deactivation of the TSX functionality.) 5086 5087 auto - Disable TSX if X86_BUG_TAA is present, 5088 otherwise enable TSX on the system. 5089 5090 Not specifying this option is equivalent to tsx=off. 5091 5092 See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst 5093 for more details. 5094 5095 tsx_async_abort= [X86,INTEL] Control mitigation for the TSX Async 5096 Abort (TAA) vulnerability. 5097 5098 Similar to Micro-architectural Data Sampling (MDS) 5099 certain CPUs that support Transactional 5100 Synchronization Extensions (TSX) are vulnerable to an 5101 exploit against CPU internal buffers which can forward 5102 information to a disclosure gadget under certain 5103 conditions. 5104 5105 In vulnerable processors, the speculatively forwarded 5106 data can be used in a cache side channel attack, to 5107 access data to which the attacker does not have direct 5108 access. 5109 5110 This parameter controls the TAA mitigation. The 5111 options are: 5112 5113 full - Enable TAA mitigation on vulnerable CPUs 5114 if TSX is enabled. 5115 5116 full,nosmt - Enable TAA mitigation and disable SMT on 5117 vulnerable CPUs. If TSX is disabled, SMT 5118 is not disabled because CPU is not 5119 vulnerable to cross-thread TAA attacks. 5120 off - Unconditionally disable TAA mitigation 5121 5122 On MDS-affected machines, tsx_async_abort=off can be 5123 prevented by an active MDS mitigation as both vulnerabilities 5124 are mitigated with the same mechanism so in order to disable 5125 this mitigation, you need to specify mds=off too. 5126 5127 Not specifying this option is equivalent to 5128 tsx_async_abort=full. On CPUs which are MDS affected 5129 and deploy MDS mitigation, TAA mitigation is not 5130 required and doesn't provide any additional 5131 mitigation. 5132 5133 For details see: 5134 Documentation/admin-guide/hw-vuln/tsx_async_abort.rst 5135 5136 turbografx.map[2|3]= [HW,JOY] 5137 TurboGraFX parallel port interface 5138 Format: 5139 <port#>,<js1>,<js2>,<js3>,<js4>,<js5>,<js6>,<js7> 5140 See also Documentation/input/devices/joystick-parport.rst 5141 5142 udbg-immortal [PPC] When debugging early kernel crashes that 5143 happen after console_init() and before a proper 5144 console driver takes over, this boot options might 5145 help "seeing" what's going on. 5146 5147 uhash_entries= [KNL,NET] 5148 Set number of hash buckets for UDP/UDP-Lite connections 5149 5150 uhci-hcd.ignore_oc= 5151 [USB] Ignore overcurrent events (default N). 5152 Some badly-designed motherboards generate lots of 5153 bogus events, for ports that aren't wired to 5154 anything. Set this parameter to avoid log spamming. 5155 Note that genuine overcurrent events won't be 5156 reported either. 5157 5158 unknown_nmi_panic 5159 [X86] Cause panic on unknown NMI. 5160 5161 usbcore.authorized_default= 5162 [USB] Default USB device authorization: 5163 (default -1 = authorized except for wireless USB, 5164 0 = not authorized, 1 = authorized, 2 = authorized 5165 if device connected to internal port) 5166 5167 usbcore.autosuspend= 5168 [USB] The autosuspend time delay (in seconds) used 5169 for newly-detected USB devices (default 2). This 5170 is the time required before an idle device will be 5171 autosuspended. Devices for which the delay is set 5172 to a negative value won't be autosuspended at all. 5173 5174 usbcore.usbfs_snoop= 5175 [USB] Set to log all usbfs traffic (default 0 = off). 5176 5177 usbcore.usbfs_snoop_max= 5178 [USB] Maximum number of bytes to snoop in each URB 5179 (default = 65536). 5180 5181 usbcore.blinkenlights= 5182 [USB] Set to cycle leds on hubs (default 0 = off). 5183 5184 usbcore.old_scheme_first= 5185 [USB] Start with the old device initialization 5186 scheme, applies only to low and full-speed devices 5187 (default 0 = off). 5188 5189 usbcore.usbfs_memory_mb= 5190 [USB] Memory limit (in MB) for buffers allocated by 5191 usbfs (default = 16, 0 = max = 2047). 5192 5193 usbcore.use_both_schemes= 5194 [USB] Try the other device initialization scheme 5195 if the first one fails (default 1 = enabled). 5196 5197 usbcore.initial_descriptor_timeout= 5198 [USB] Specifies timeout for the initial 64-byte 5199 USB_REQ_GET_DESCRIPTOR request in milliseconds 5200 (default 5000 = 5.0 seconds). 5201 5202 usbcore.nousb [USB] Disable the USB subsystem 5203 5204 usbcore.quirks= 5205 [USB] A list of quirk entries to augment the built-in 5206 usb core quirk list. List entries are separated by 5207 commas. Each entry has the form 5208 VendorID:ProductID:Flags. The IDs are 4-digit hex 5209 numbers and Flags is a set of letters. Each letter 5210 will change the built-in quirk; setting it if it is 5211 clear and clearing it if it is set. The letters have 5212 the following meanings: 5213 a = USB_QUIRK_STRING_FETCH_255 (string 5214 descriptors must not be fetched using 5215 a 255-byte read); 5216 b = USB_QUIRK_RESET_RESUME (device can't resume 5217 correctly so reset it instead); 5218 c = USB_QUIRK_NO_SET_INTF (device can't handle 5219 Set-Interface requests); 5220 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't 5221 handle its Configuration or Interface 5222 strings); 5223 e = USB_QUIRK_RESET (device can't be reset 5224 (e.g morph devices), don't use reset); 5225 f = USB_QUIRK_HONOR_BNUMINTERFACES (device has 5226 more interface descriptions than the 5227 bNumInterfaces count, and can't handle 5228 talking to these interfaces); 5229 g = USB_QUIRK_DELAY_INIT (device needs a pause 5230 during initialization, after we read 5231 the device descriptor); 5232 h = USB_QUIRK_LINEAR_UFRAME_INTR_BINTERVAL (For 5233 high speed and super speed interrupt 5234 endpoints, the USB 2.0 and USB 3.0 spec 5235 require the interval in microframes (1 5236 microframe = 125 microseconds) to be 5237 calculated as interval = 2 ^ 5238 (bInterval-1). 5239 Devices with this quirk report their 5240 bInterval as the result of this 5241 calculation instead of the exponent 5242 variable used in the calculation); 5243 i = USB_QUIRK_DEVICE_QUALIFIER (device can't 5244 handle device_qualifier descriptor 5245 requests); 5246 j = USB_QUIRK_IGNORE_REMOTE_WAKEUP (device 5247 generates spurious wakeup, ignore 5248 remote wakeup capability); 5249 k = USB_QUIRK_NO_LPM (device can't handle Link 5250 Power Management); 5251 l = USB_QUIRK_LINEAR_FRAME_INTR_BINTERVAL 5252 (Device reports its bInterval as linear 5253 frames instead of the USB 2.0 5254 calculation); 5255 m = USB_QUIRK_DISCONNECT_SUSPEND (Device needs 5256 to be disconnected before suspend to 5257 prevent spurious wakeup); 5258 n = USB_QUIRK_DELAY_CTRL_MSG (Device needs a 5259 pause after every control message); 5260 o = USB_QUIRK_HUB_SLOW_RESET (Hub needs extra 5261 delay after resetting its port); 5262 Example: quirks=0781:5580:bk,0a5c:5834:gij 5263 5264 usbhid.mousepoll= 5265 [USBHID] The interval which mice are to be polled at. 5266 5267 usbhid.jspoll= 5268 [USBHID] The interval which joysticks are to be polled at. 5269 5270 usbhid.kbpoll= 5271 [USBHID] The interval which keyboards are to be polled at. 5272 5273 usb-storage.delay_use= 5274 [UMS] The delay in seconds before a new device is 5275 scanned for Logical Units (default 1). 5276 5277 usb-storage.quirks= 5278 [UMS] A list of quirks entries to supplement or 5279 override the built-in unusual_devs list. List 5280 entries are separated by commas. Each entry has 5281 the form VID:PID:Flags where VID and PID are Vendor 5282 and Product ID values (4-digit hex numbers) and 5283 Flags is a set of characters, each corresponding 5284 to a common usb-storage quirk flag as follows: 5285 a = SANE_SENSE (collect more than 18 bytes 5286 of sense data, not on uas); 5287 b = BAD_SENSE (don't collect more than 18 5288 bytes of sense data, not on uas); 5289 c = FIX_CAPACITY (decrease the reported 5290 device capacity by one sector); 5291 d = NO_READ_DISC_INFO (don't use 5292 READ_DISC_INFO command, not on uas); 5293 e = NO_READ_CAPACITY_16 (don't use 5294 READ_CAPACITY_16 command); 5295 f = NO_REPORT_OPCODES (don't use report opcodes 5296 command, uas only); 5297 g = MAX_SECTORS_240 (don't transfer more than 5298 240 sectors at a time, uas only); 5299 h = CAPACITY_HEURISTICS (decrease the 5300 reported device capacity by one 5301 sector if the number is odd); 5302 i = IGNORE_DEVICE (don't bind to this 5303 device); 5304 j = NO_REPORT_LUNS (don't use report luns 5305 command, uas only); 5306 l = NOT_LOCKABLE (don't try to lock and 5307 unlock ejectable media, not on uas); 5308 m = MAX_SECTORS_64 (don't transfer more 5309 than 64 sectors = 32 KB at a time, 5310 not on uas); 5311 n = INITIAL_READ10 (force a retry of the 5312 initial READ(10) command, not on uas); 5313 o = CAPACITY_OK (accept the capacity 5314 reported by the device, not on uas); 5315 p = WRITE_CACHE (the device cache is ON 5316 by default, not on uas); 5317 r = IGNORE_RESIDUE (the device reports 5318 bogus residue values, not on uas); 5319 s = SINGLE_LUN (the device has only one 5320 Logical Unit); 5321 t = NO_ATA_1X (don't allow ATA(12) and ATA(16) 5322 commands, uas only); 5323 u = IGNORE_UAS (don't bind to the uas driver); 5324 w = NO_WP_DETECT (don't test whether the 5325 medium is write-protected). 5326 y = ALWAYS_SYNC (issue a SYNCHRONIZE_CACHE 5327 even if the device claims no cache, 5328 not on uas) 5329 Example: quirks=0419:aaf5:rl,0421:0433:rc 5330 5331 user_debug= [KNL,ARM] 5332 Format: <int> 5333 See arch/arm/Kconfig.debug help text. 5334 1 - undefined instruction events 5335 2 - system calls 5336 4 - invalid data aborts 5337 8 - SIGSEGV faults 5338 16 - SIGBUS faults 5339 Example: user_debug=31 5340 5341 userpte= 5342 [X86] Flags controlling user PTE allocations. 5343 5344 nohigh = do not allocate PTE pages in 5345 HIGHMEM regardless of setting 5346 of CONFIG_HIGHPTE. 5347 5348 vdso= [X86,SH] 5349 On X86_32, this is an alias for vdso32=. Otherwise: 5350 5351 vdso=1: enable VDSO (the default) 5352 vdso=0: disable VDSO mapping 5353 5354 vdso32= [X86] Control the 32-bit vDSO 5355 vdso32=1: enable 32-bit VDSO 5356 vdso32=0 or vdso32=2: disable 32-bit VDSO 5357 5358 See the help text for CONFIG_COMPAT_VDSO for more 5359 details. If CONFIG_COMPAT_VDSO is set, the default is 5360 vdso32=0; otherwise, the default is vdso32=1. 5361 5362 For compatibility with older kernels, vdso32=2 is an 5363 alias for vdso32=0. 5364 5365 Try vdso32=0 if you encounter an error that says: 5366 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed! 5367 5368 vector= [IA-64,SMP] 5369 vector=percpu: enable percpu vector domain 5370 5371 video= [FB] Frame buffer configuration 5372 See Documentation/fb/modedb.rst. 5373 5374 video.brightness_switch_enabled= [0,1] 5375 If set to 1, on receiving an ACPI notify event 5376 generated by hotkey, video driver will adjust brightness 5377 level and then send out the event to user space through 5378 the allocated input device; If set to 0, video driver 5379 will only send out the event without touching backlight 5380 brightness level. 5381 default: 1 5382 5383 virtio_mmio.device= 5384 [VMMIO] Memory mapped virtio (platform) device. 5385 5386 <size>@<baseaddr>:<irq>[:<id>] 5387 where: 5388 <size> := size (can use standard suffixes 5389 like K, M and G) 5390 <baseaddr> := physical base address 5391 <irq> := interrupt number (as passed to 5392 request_irq()) 5393 <id> := (optional) platform device id 5394 example: 5395 virtio_mmio.device=1K@0x100b0000:48:7 5396 5397 Can be used multiple times for multiple devices. 5398 5399 vga= [BOOT,X86-32] Select a particular video mode 5400 See Documentation/x86/boot.rst and 5401 Documentation/admin-guide/svga.rst. 5402 Use vga=ask for menu. 5403 This is actually a boot loader parameter; the value is 5404 passed to the kernel using a special protocol. 5405 5406 vm_debug[=options] [KNL] Available with CONFIG_DEBUG_VM=y. 5407 May slow down system boot speed, especially when 5408 enabled on systems with a large amount of memory. 5409 All options are enabled by default, and this 5410 interface is meant to allow for selectively 5411 enabling or disabling specific virtual memory 5412 debugging features. 5413 5414 Available options are: 5415 P Enable page structure init time poisoning 5416 - Disable all of the above options 5417 5418 vmalloc=nn[KMG] [KNL,BOOT] Forces the vmalloc area to have an exact 5419 size of <nn>. This can be used to increase the 5420 minimum size (128MB on x86). It can also be used to 5421 decrease the size and leave more room for directly 5422 mapped kernel RAM. 5423 5424 vmcp_cma=nn[MG] [KNL,S390] 5425 Sets the memory size reserved for contiguous memory 5426 allocations for the vmcp device driver. 5427 5428 vmhalt= [KNL,S390] Perform z/VM CP command after system halt. 5429 Format: <command> 5430 5431 vmpanic= [KNL,S390] Perform z/VM CP command after kernel panic. 5432 Format: <command> 5433 5434 vmpoff= [KNL,S390] Perform z/VM CP command after power off. 5435 Format: <command> 5436 5437 vsyscall= [X86-64] 5438 Controls the behavior of vsyscalls (i.e. calls to 5439 fixed addresses of 0xffffffffff600x00 from legacy 5440 code). Most statically-linked binaries and older 5441 versions of glibc use these calls. Because these 5442 functions are at fixed addresses, they make nice 5443 targets for exploits that can control RIP. 5444 5445 emulate [default] Vsyscalls turn into traps and are 5446 emulated reasonably safely. The vsyscall 5447 page is readable. 5448 5449 xonly Vsyscalls turn into traps and are 5450 emulated reasonably safely. The vsyscall 5451 page is not readable. 5452 5453 none Vsyscalls don't work at all. This makes 5454 them quite hard to use for exploits but 5455 might break your system. 5456 5457 vt.color= [VT] Default text color. 5458 Format: 0xYX, X = foreground, Y = background. 5459 Default: 0x07 = light gray on black. 5460 5461 vt.cur_default= [VT] Default cursor shape. 5462 Format: 0xCCBBAA, where AA, BB, and CC are the same as 5463 the parameters of the <Esc>[?A;B;Cc escape sequence; 5464 see VGA-softcursor.txt. Default: 2 = underline. 5465 5466 vt.default_blu= [VT] 5467 Format: <blue0>,<blue1>,<blue2>,...,<blue15> 5468 Change the default blue palette of the console. 5469 This is a 16-member array composed of values 5470 ranging from 0-255. 5471 5472 vt.default_grn= [VT] 5473 Format: <green0>,<green1>,<green2>,...,<green15> 5474 Change the default green palette of the console. 5475 This is a 16-member array composed of values 5476 ranging from 0-255. 5477 5478 vt.default_red= [VT] 5479 Format: <red0>,<red1>,<red2>,...,<red15> 5480 Change the default red palette of the console. 5481 This is a 16-member array composed of values 5482 ranging from 0-255. 5483 5484 vt.default_utf8= 5485 [VT] 5486 Format=<0|1> 5487 Set system-wide default UTF-8 mode for all tty's. 5488 Default is 1, i.e. UTF-8 mode is enabled for all 5489 newly opened terminals. 5490 5491 vt.global_cursor_default= 5492 [VT] 5493 Format=<-1|0|1> 5494 Set system-wide default for whether a cursor 5495 is shown on new VTs. Default is -1, 5496 i.e. cursors will be created by default unless 5497 overridden by individual drivers. 0 will hide 5498 cursors, 1 will display them. 5499 5500 vt.italic= [VT] Default color for italic text; 0-15. 5501 Default: 2 = green. 5502 5503 vt.underline= [VT] Default color for underlined text; 0-15. 5504 Default: 3 = cyan. 5505 5506 watchdog timers [HW,WDT] For information on watchdog timers, 5507 see Documentation/watchdog/watchdog-parameters.rst 5508 or other driver-specific files in the 5509 Documentation/watchdog/ directory. 5510 5511 watchdog_thresh= 5512 [KNL] 5513 Set the hard lockup detector stall duration 5514 threshold in seconds. The soft lockup detector 5515 threshold is set to twice the value. A value of 0 5516 disables both lockup detectors. Default is 10 5517 seconds. 5518 5519 workqueue.watchdog_thresh= 5520 If CONFIG_WQ_WATCHDOG is configured, workqueue can 5521 warn stall conditions and dump internal state to 5522 help debugging. 0 disables workqueue stall 5523 detection; otherwise, it's the stall threshold 5524 duration in seconds. The default value is 30 and 5525 it can be updated at runtime by writing to the 5526 corresponding sysfs file. 5527 5528 workqueue.disable_numa 5529 By default, all work items queued to unbound 5530 workqueues are affine to the NUMA nodes they're 5531 issued on, which results in better behavior in 5532 general. If NUMA affinity needs to be disabled for 5533 whatever reason, this option can be used. Note 5534 that this also can be controlled per-workqueue for 5535 workqueues visible under /sys/bus/workqueue/. 5536 5537 workqueue.power_efficient 5538 Per-cpu workqueues are generally preferred because 5539 they show better performance thanks to cache 5540 locality; unfortunately, per-cpu workqueues tend to 5541 be more power hungry than unbound workqueues. 5542 5543 Enabling this makes the per-cpu workqueues which 5544 were observed to contribute significantly to power 5545 consumption unbound, leading to measurably lower 5546 power usage at the cost of small performance 5547 overhead. 5548 5549 The default value of this parameter is determined by 5550 the config option CONFIG_WQ_POWER_EFFICIENT_DEFAULT. 5551 5552 workqueue.debug_force_rr_cpu 5553 Workqueue used to implicitly guarantee that work 5554 items queued without explicit CPU specified are put 5555 on the local CPU. This guarantee is no longer true 5556 and while local CPU is still preferred work items 5557 may be put on foreign CPUs. This debug option 5558 forces round-robin CPU selection to flush out 5559 usages which depend on the now broken guarantee. 5560 When enabled, memory and cache locality will be 5561 impacted. 5562 5563 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of 5564 default x2apic cluster mode on platforms 5565 supporting x2apic. 5566 5567 x86_intel_mid_timer= [X86-32,APBT] 5568 Choose timer option for x86 Intel MID platform. 5569 Two valid options are apbt timer only and lapic timer 5570 plus one apbt timer for broadcast timer. 5571 x86_intel_mid_timer=apbt_only | lapic_and_apbt 5572 5573 xen_512gb_limit [KNL,X86-64,XEN] 5574 Restricts the kernel running paravirtualized under Xen 5575 to use only up to 512 GB of RAM. The reason to do so is 5576 crash analysis tools and Xen tools for doing domain 5577 save/restore/migration must be enabled to handle larger 5578 domains. 5579 5580 xen_emul_unplug= [HW,X86,XEN] 5581 Unplug Xen emulated devices 5582 Format: [unplug0,][unplug1] 5583 ide-disks -- unplug primary master IDE devices 5584 aux-ide-disks -- unplug non-primary-master IDE devices 5585 nics -- unplug network devices 5586 all -- unplug all emulated devices (NICs and IDE disks) 5587 unnecessary -- unplugging emulated devices is 5588 unnecessary even if the host did not respond to 5589 the unplug protocol 5590 never -- do not unplug even if version check succeeds 5591 5592 xen_legacy_crash [X86,XEN] 5593 Crash from Xen panic notifier, without executing late 5594 panic() code such as dumping handler. 5595 5596 xen_nopvspin [X86,XEN] 5597 Disables the ticketlock slowpath using Xen PV 5598 optimizations. 5599 5600 xen_nopv [X86] 5601 Disables the PV optimizations forcing the HVM guest to 5602 run as generic HVM guest with no PV drivers. 5603 This option is obsoleted by the "nopv" option, which 5604 has equivalent effect for XEN platform. 5605 5606 xen_scrub_pages= [XEN] 5607 Boolean option to control scrubbing pages before giving them back 5608 to Xen, for use by other domains. Can be also changed at runtime 5609 with /sys/devices/system/xen_memory/xen_memory0/scrub_pages. 5610 Default value controlled with CONFIG_XEN_SCRUB_PAGES_DEFAULT. 5611 5612 xen_timer_slop= [X86-64,XEN] 5613 Set the timer slop (in nanoseconds) for the virtual Xen 5614 timers (default is 100000). This adjusts the minimum 5615 delta of virtualized Xen timers, where lower values 5616 improve timer resolution at the expense of processing 5617 more timer interrupts. 5618 5619 nopv= [X86,XEN,KVM,HYPER_V,VMWARE] 5620 Disables the PV optimizations forcing the guest to run 5621 as generic guest with no PV drivers. Currently support 5622 XEN HVM, KVM, HYPER_V and VMWARE guest. 5623 5624 xirc2ps_cs= [NET,PCMCIA] 5625 Format: 5626 <irq>,<irq_mask>,<io>,<full_duplex>,<do_sound>,<lockup_hack>[,<irq2>[,<irq3>[,<irq4>]]] 5627 5628 xive= [PPC] 5629 By default on POWER9 and above, the kernel will 5630 natively use the XIVE interrupt controller. This option 5631 allows the fallback firmware mode to be used: 5632 5633 off Fallback to firmware control of XIVE interrupt 5634 controller on both pseries and powernv 5635 platforms. Only useful on POWER9 and above. 5636 5637 xhci-hcd.quirks [USB,KNL] 5638 A hex value specifying bitmask with supplemental xhci 5639 host controller quirks. Meaning of each bit can be 5640 consulted in header drivers/usb/host/xhci.h. 5641 5642 xmon [PPC] 5643 Format: { early | on | rw | ro | off } 5644 Controls if xmon debugger is enabled. Default is off. 5645 Passing only "xmon" is equivalent to "xmon=early". 5646 early Call xmon as early as possible on boot; xmon 5647 debugger is called from setup_arch(). 5648 on xmon debugger hooks will be installed so xmon 5649 is only called on a kernel crash. Default mode, 5650 i.e. either "ro" or "rw" mode, is controlled 5651 with CONFIG_XMON_DEFAULT_RO_MODE. 5652 rw xmon debugger hooks will be installed so xmon 5653 is called only on a kernel crash, mode is write, 5654 meaning SPR registers, memory and, other data 5655 can be written using xmon commands. 5656 ro same as "rw" option above but SPR registers, 5657 memory, and other data can't be written using 5658 xmon commands. 5659 off xmon is disabled. 5660