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