Lines Matching +full:can +full:- +full:disable
9 accept_memory=eager can be used to accept all memory
16 force -- enable ACPI if default was off
17 on -- enable ACPI but allow fallback to DT [arm64,riscv64]
18 off -- disable ACPI if default was on
19 noirq -- do not use ACPI for IRQ routing
20 strict -- Be less tolerant of platforms that are not
22 rsdt -- prefer RSDT over (default) XSDT
23 copy_dsdt -- copy DSDT to memory
24 nocmcff -- Disable firmware first mode for corrected
28 nospcr -- disable console in ACPI SPCR table as
45 If set to vendor, prefer vendor-specific driver
50 If set to none, disable the ACPI backlight interface.
59 Disable AML predefined validation mechanism
60 This mechanism can repair the evaluation result to make
77 Documentation/firmware-guide/acpi/debug.rst for more information about
98 can interfere with legacy drivers.
130 by unsupported hardware/firmware features can result in
133 This facility can be used to prevent such uncontrolled
135 Format: <byte> or <bitmap-list>
138 Disable auto-serialization of AML methods
141 auto-serialization feature.
145 acpi_no_memhotplug [ACPI] Disable memory hotplug. Useful for kdump
149 Disable installation of static SSDTs at early boot time
159 Ignore the ACPI-based watchdog interface (WDAT) and let
180 acpi_osi=! # disable all built-in OS vendor
182 acpi_osi=!! # enable all built-in OS vendor
184 acpi_osi= # disable all strings
186 'acpi_osi=!' can be used in combination with single or
188 vendor string(s). Note that such command can only
199 can make '_OSI("Windows 2000")' TRUE.
203 exist in the ACPI namespace. NOTE that such command can
208 1. 'acpi_osi=' can make 'CondRefOf(_OSI, Local1)'
211 'acpi_osi=!*' can be used in combination with single or
213 string(s). Note that such command can affect the
223 1. 'acpi_osi="Module Device" acpi_osi=!*' can make
225 2. 'acpi_osi=!* acpi_osi="Module Device"' can make
244 For broken nForce2 BIOS resulting in XT-PIC timer.
253 as soon as the kernel's real-mode entry point is called.
286 off: disable AGP support
291 See Documentation/sound/alsa-configuration.rst
298 align_va_addr= [X86-64]
303 CPU-intensive style benchmark, and it can vary highly in
306 32: only for 32-bit processes
307 64: only for 64-bit processes
308 on: enable for both 32- and 64-bit processes
309 off: disable for both 32- and 64-bit processes
319 Allow execve() of 32-bit applications and setting of the
321 subset of the CPUs support 32-bit EL0. When this
322 parameter is present, the set of CPUs supporting 32-bit
324 and hot-unplug operations may be restricted.
326 See Documentation/arch/arm64/asymmetric-32bit.rst for more
329 amd_iommu= [HW,X86-64]
332 fullflush - Deprecated, equivalent to iommu.strict=1
333 off - do not initialize any AMD IOMMU found in
335 force_isolation - Force device isolation for all
340 force_enable - Force enable the IOMMU on platforms known
343 pgtbl_v1 - Use v1 page table for DMA-API (Default).
344 pgtbl_v2 - Use v2 page table for DMA-API.
345 irtcachedis - Disable Interrupt Remapping Table (IRT) caching.
346 nohugepages - Limit page-sizes used for v1 page-tables
348 v2_pgsizes_only - Limit page-sizes used for v1 page-tables
352 amd_iommu_dump= [HW,X86-64]
358 amd_iommu_intr= [HW,X86-64]
361 legacy - Use legacy interrupt remapping mode.
362 vapic - Use virtual APIC mode, which allows IOMMU
364 This mode requires kvm-amd.avic=1.
368 disable
392 disable
393 Disable amd-pstate preferred core.
406 Power management functions (SPARCstation-4/5 + deriv.)
408 Disable APC CPU standby support. SPARCstation-Fox does
409 not play well with APC CPU idle - disable it if you have
412 apic [APIC,X86-64] Use IO-APIC. Default.
418 when initialising the APIC and IO-APIC components.
439 apparmor= [APPARMOR] Disable or enable AppArmor at boot time
442 0 -- disable.
443 1 -- enable.
446 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards
449 arm64.no32bit_el0 [ARM64] Unconditionally disable the execution of
452 arm64.nobti [ARM64] Unconditionally disable Branch Target
455 arm64.nogcs [ARM64] Unconditionally disable Guarded Control Stack
458 arm64.nomops [ARM64] Unconditionally disable Memory Copy and Memory
461 arm64.nomte [ARM64] Unconditionally disable Memory Tagging Extension
464 arm64.nopauth [ARM64] Unconditionally disable Pointer Authentication
467 arm64.nosme [ARM64] Unconditionally disable Scalable Matrix
470 arm64.nosve [ARM64] Unconditionally disable Scalable Vector
494 audit= [KNL] Enable the audit sub-system
496 0 | off - kernel audit is disabled and can not be
498 unset - kernel audit is initialized but disabled and
500 1 | on - kernel audit is initialized and partially
511 behavior is to disable the BAU (i.e. bau=0).
513 0 - Disable the BAU.
514 1 - Enable the BAU.
515 unset - Disable the BAU.
543 O_EXCL. Default is Y and can be changed through the
547 Disable BERT OS support on buggy BIOSes.
550 Disable BGRT to avoid flickering OEM logo.
554 See Documentation/block/cmdline-partition.rst
565 Extended command line options can be added to an initrd
568 See Documentation/admin-guide/bootconfig.rst
573 bttv.pll= See Documentation/admin-guide/media/bttv.rst
582 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection.
601 inclusive. See arch/mips/include/asm/pgtable-bits.h
608 cgroup_disable= [KNL] Disable a particular controller or optional feature
609 Format: {name of the controller(s) or feature(s) to disable}
611 - foo isn't auto-mounted if you mount all cgroups in
613 - foo isn't visible as an individually mountable
615 - if foo is an optional feature then the feature is
619 cut the overhead, others just disable the usage. So
621 Specifying "pressure" disables per-cgroup pressure
624 cgroup_no_v1= [KNL] Disable cgroup controllers and named hierarchies in v1
633 cgroup_favordynmods= [KNL] Enable or Disable favordynmods.
639 nosocket -- Disable socket memory accounting.
640 nokmem -- Disable kernel memory accounting.
641 nobpf -- Disable BPF memory accounting.
646 0 -- check protection applied by kernel (includes
648 1 -- check protection requested by application.
650 Value can be changed at runtime via
658 Disable CPUID feature X for the kernel. See
660 numbers X. Note the Linux-specific bits are not necessarily
661 stable over kernel options, but the vendor-specific
663 X can also be a string as appearing in the flags: line
672 Also note the kernel might malfunction if you disable
681 force such clocks to be always-on nor does it reserve
685 information, see Documentation/driver-api/clk.rst.
687 clock= [BUGS=X86-32, HW] gettimeofday clocksource override.
704 [X86-32] pit,hpet,tsc;
711 [X86-64] hpet,tsc
716 Enable/disable the eventstream feature of the ARM
717 architected timer so that code using WFE-based polling
718 loops can be debugged more effectively on production
731 clocksource-wdtest.holdoff= [KNL]
737 cma=nn[MG]@[start[MG][-end[MG]]]
748 Sets the size of kernel per-numa memory area for
750 per-numa CMA altogether. And If this option is not
752 With per-numa CMA enabled, DMA users on node nid will
779 com20020= [HW,NET] ARCnet - COM20020 chipset
783 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers)
787 ARCnet - COM90xx chipset (memory-mapped buffers)
816 See Documentation/admin-guide/serial-console.rst for more
829 can be viewed with:
831 $ ls -d /sys/bus/serial-base/devices/*:*.*/tty/*
832 /sys/bus/serial-base/devices/00:04:0.0/tty/ttyS0
834 In the above example, the console can be addressed with
845 Start an early, polled-mode console on the 8250/16550
848 MMIO inter-register address stride is either 8-bit
849 (mmio), 16-bit (mmio16), or 32-bit (mmio32).
853 the h/w is not re-initialized.
859 Use to disable console output, i.e., to have kernel
880 syscall, or to executing "dmesg -S --raw" or to reading
895 Enable/disable the CPU sampling based debugging.
896 0: default value, disable debugging
904 disable the cpuidle sub-system
910 disable the cpufreq sub-system
918 [X86,EARLY] Delay for N microsec between assert and de-assert
924 [SMP] Enable/disable parallel bringup of secondary CPUs
933 notifiers can make the crashed kernel more unstable.
939 like Hyper-V, PowerPC (fadump) and AMD SEV-SNP.
942 [KNL,EARLY] Using kexec, Linux can switch to a 'crash kernel'
947 [KNL, X86-64, ARM64, RISCV, LoongArch] Select a region
950 See Documentation/admin-guide/kdump/kdump.rst for further details.
955 start-[end] where start and end are both
957 Documentation/admin-guide/kdump/kdump.rst for an example.
960 [KNL, X86-64, ARM64, RISCV, LoongArch] range could be
968 [KNL, X86-64, ARM64, RISCV, LoongArch] range under 4G.
974 for 32-bit devices won't run out. Kernel would try to allocate
977 --> x86: max(swiotlb_size_or_default() + 8MiB, 256MiB)
978 --> arm64: 128MiB
979 --> riscv: 128MiB
980 --> loongarch: 128MiB
983 0: to disable low allocation.
988 [KNL] Disable crypto self-tests
996 csdlock_debug= [KNL] Enable or disable debug add-ons of cross-CPU
1011 See also Documentation/input/devices/joystick-parport.rst
1024 [KNL] verbose locking self-tests
1027 self-tests.
1029 (no extra messages), setting it to -1 (all bits set)
1030 will print _a_lot_ more information - normally only
1065 Format: { on, no-mount, off }
1067 no-mount:
1068 Filesystem is not registered but kernel clients can
1069 access APIs and a crashkernel can be used to read
1072 get a -EPERM as result when trying to register files
1076 Default value is set in build-time with a kernel configuration.
1088 Documentation/admin-guide/mm/hugetlbpage.rst.
1102 delayacct [KNL] Enable per-task delay accounting
1146 can be useful when debugging issues that require an SLB
1149 disable= [IPV6]
1153 Disable RADIX MMU mode on POWER9
1156 Disable TLBIE instruction. Currently does not work
1160 Disable Dynamic DMA Window support. Use this
1178 Disable PIN 1 of APIC timer
1179 Can be useful to work around chipset bugs.
1181 dis_ucode_ldr [X86] Disable the microcode loader.
1188 entries for DMA-API debugging code. One entry is
1189 required per DMA-API allocation. Use this if the
1190 DMA-API debugging code disables itself because the
1194 With this option the DMA-API debugging driver
1195 filter feature can be enabled at boot time. Just
1197 The filter can be disabled or changed to another
1214 are enabled. In order to disable RFDS mitigation all
1218 Documentation/admin-guide/hw-vuln/reg-file-data-sampling.rst
1243 Control how the dt_cpu_ftrs device-tree binding is
1253 what data is available or for reverse-engineering.
1258 Documentation/admin-guide/dynamic-debug-howto.rst
1269 determined by stdout-path property in device tree's
1274 Start an early, polled-mode console on a Cadence
1285 Start an early, polled-mode console on the 8250/16550
1287 MMIO inter-register address stride is either 8-bit
1288 (mmio) or 32-bit (mmio32 or mmio32be).
1298 Start an early, polled-mode console on a pl011 serial
1302 the driver will use only 32-bit accessors to read/write
1311 Start an early, polled-mode console on a meson serial
1317 Start an early, polled-mode console on an msm serial
1323 Start an early, polled-mode console on an msm serial
1329 Start an early, polled-mode console on a serial port
1335 Start an early, polled-mode console on a serial port
1341 Use RISC-V SBI (Supervisor Binary Interface) for early
1359 Start an early, polled-mode console on a lantiq serial
1373 Start an early, polled-mode, output-only console on the
1378 Start an early, polled-mode console on the
1384 Start an early, polled-mode console on a Qualcomm
1392 coherent non-x86 systems that use system memory for
1422 Append ",keep" to not disable it when the real console
1425 Only one of vga, serial, or usb debug port can
1433 You can find the port for a given device in
1443 The xen option can only be used in Xen domains.
1445 The sclp output can only be used on s390.
1447 The bios output can only be used on SuperH.
1457 off: disable H/W event reporting through EDAC.
1469 disable_early_pci_dma: disable the busmaster bit on all
1471 nochunk: disable reading files in "chunks" in the EFI
1472 boot stub, as chunking can cause problems with some
1474 noruntime : disable EFI runtime services support
1478 claim. Specify efi=nosoftreserve to disable this
1486 Using this parameter you can use more than 50% of
1495 Documentation/admin-guide/acpi/ssdt-overlays.rst for details.
1508 but can only be used if the backing tty is available
1512 elanfreq= [X86-32]
1520 See Documentation/admin-guide/kdump/kdump.rst for details.
1529 Can be useful to work around chipset bugs
1536 0 -- permissive (log only, no denials).
1537 1 -- enforcing (deny and log).
1539 Value can be changed at runtime via
1543 Disable Error Record Serialization Table (ERST)
1570 See also Documentation/fault-injection/.
1574 See Documentation/admin-guide/sysctl/net.rst for
1578 See Documentation/admin-guide/blockdev/floppy.rst.
1580 forcepae [X86-32]
1582 Many Pentium M systems disable PAE but may have a
1587 fred= [X86-64]
1588 Enable/disable Flexible Return and Event Delivery.
1591 off: disable FRED, the default setting.
1600 ftrace ring buffer that can be read at:
1606 Optionally, the snapshot can also be defined for a tracing
1638 ftrace_filter=[function-list]
1640 tracer at boot up. function-list is a comma-separated
1641 list of functions. This list can be changed at run
1645 ftrace_notrace=[function-list]
1647 function-list. This list can be changed at run time
1651 ftrace_graph_filter=[function-list]
1654 function-list is a comma-separated list of functions
1655 that can be changed at run time by the
1658 ftrace_graph_notrace=[function-list]
1660 function-list. This list is a comma-separated list of
1661 functions that can be changed at run time by the
1667 can be changed at run time by the max_graph_depth file
1680 off -- Don't create device links from firmware info.
1681 permissive -- Create device links from firmware info
1684 on -- Create device links from firmware info and use it
1686 rpm -- Like "on", but also use to order runtime PM.
1699 strict -- Default. Continue waiting on consumers to
1701 timeout -- Give up waiting on consumers and call
1711 See also Documentation/input/devices/joystick-parport.rst
1715 gart_fix_e820= [X86-64,EARLY] disable the fix e820 for K8 GART
1728 The mitigation may have a performance impact but can be
1732 force: Disable AVX to mitigate systems without
1737 off: Disable GDS mitigation.
1741 gcov_persist= [GCOV] When non-zero (default), profiling data for
1751 gpio-mockup.gpio_mockup_ranges
1754 gpio-mockup.gpio_mockup_named_lines
1786 against bounds-checking flaws in the kernel's
1791 off Disable hardened usercopy checks.
1794 [KNL] Should the hard-lockup detector generate
1800 for 64-bit NUMA, off otherwise.
1807 Disable Hardware Error Source Table (HEST) support;
1808 corresponding firmware-first mode error processing
1815 no Disable hibernation and resume.
1818 during restoration read-only).
1843 highres= [KNL] Enable/disable high resolution timer mode.
1862 hpet= [X86-32,HPET] option to control HPET usage
1863 Format: { enable (default) | disable | force |
1865 disable: disable HPET and use PIT instead
1879 number of pages to allocate per-node can be specified.
1880 See also Documentation/admin-guide/mm/hugetlbpage.rst.
1888 pair hugepagesz=X hugepages=Y can be specified once
1891 Documentation/admin-guide/mm/hugetlbpage.rst.
1896 allocate hugepages during boot. This option can be
1901 Note that this parameter only applies to non-gigantic huge pages.
1905 of a CMA area per node can be specified.
1911 boot-time allocation of gigantic hugepages is skipped.
1929 off: disable HVO
1946 by the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time
1947 option. The value selected by this boot parameter can
1952 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1974 Set device-properties for ACPI-enumerated I2C-attached
1975 touchscreen, to e.g. fix coordinates of upside-down
1983 Omit "=<val>" entirely Set a boolean device-property
1984 Unsigned number Set a u32 device-property
1985 Anything else Set a string device-property
1988 i2c_touchscreen_props=GDIX1001:touchscreen-inverted-x:
1989 touchscreen-inverted-y
1991 i2c_touchscreen_props=MSSL1680:touchscreen-size-x=1920:
1992 touchscreen-size-y=1080:touchscreen-inverted-y:
1993 firmware-name=gsl1680-vendor-model.fw:silead,home-button
1998 (disabled by default, and as a pre-condition
2000 i8042.direct [HW] Put keyboard port into non-translated mode
2001 i8042.dumbkbd [HW] Pretend that controller can only read data from
2006 i8042.noloop [HW] Disable the AUX Loopback command while probing
2014 suspend-to-ram transitions, only during s2r
2038 -1 -- never invert brightness
2039 0 -- machine default
2040 1 -- force brightness inversion
2042 ia32_emulation= [X86-64]
2044 When true, allows loading 32-bit programs and executing 32-bit
2069 idle=nomwait: Disable mwait for CPU C-states
2097 legacy only accept legacy-NaN binaries, if supported
2099 2008 only accept 2008-NaN binaries, if supported
2110 'relaxed' straps the emulator for both legacy-NaN and
2111 2008-NaN, whereas 'strict' enables legacy-NaN only on
2120 Ignore loglevel setting - this will print /all/
2128 print warning at first misuse. Can be changed via
2187 Formats: { "ima" | "ima-ng" | "ima-ngv2" | "ima-sig" |
2188 "ima-sigv2" }
2189 Default: "ima-ng"
2201 different crypto accelerators. This option can be used
2209 different crypto accelerators. This option can be used
2217 off: Disable mitigation.
2222 stuff: Deploy RSB-fill mitigation when retpoline is
2227 Documentation/admin-guide/hw-vuln/indirect-target-selection.rst
2238 initcall_blacklist= [KNL] Do not execute a comma-separated list of
2239 initcall functions. Useful for debugging built-in
2249 but as a debugging aid, one can get the
2274 default (disallow access to all but pkey 0). Can
2284 0 -- basic integrity auditing messages. (Default)
2285 1 -- additional integrity auditing messages.
2291 Disable intel iommu driver.
2313 could harm performance of some high-throughput
2322 1 to 9 specify maximum depth of C-state.
2325 disable
2331 algorithms for p-state selection. There are two
2332 P-state selection algorithms provided by
2335 on whether or not the hardware managed P-states
2342 used along with the hardware-managed P-states (HWP)
2346 in favor of acpi-cpufreq. Forcing the intel_pstate driver
2347 instead of acpi-cpufreq may disable platform features, such
2349 P-States information being indicated to OSPM and therefore
2352 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
2365 Allow per-logical-CPU P-State performance control limits using
2368 Do not enable capacity-aware scheduling (CAS) on
2371 intremap= [X86-64,Intel-IOMMU,EARLY]
2373 off disable Interrupt Remapping
2374 nosid disable Source ID checking
2376 BIOS x2APIC opt-out request will be ignored
2377 nopost disable Interrupt Posting
2381 iomem= Disable strict checking of access to MMIO memory
2407 Intel machines). This can be used to prevent the usage
2416 Disable IOMMU bypass, using IOMMU for PCI devices.
2419 AMD Gart HW IOMMU-specific options:
2438 Do scatter-gather (SG) merging. Implies "force"
2442 Don't do scatter-gather (SG) merging.
2455 0 - Try to allocate a 32-bit DMA address first, before
2457 1 - Allocate directly from the full usable range,
2459 greater than 32-bit addressing.
2463 0 - Lazy mode.
2469 1 - Strict mode.
2472 unset - Use value of CONFIG_IOMMU_DEFAULT_DMA_{LAZY,STRICT}.
2474 legacy driver-specific options takes precedence.
2479 0 - Use IOMMU translation for DMA.
2480 1 - Bypass the IOMMU for DMA.
2481 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH.
2483 io7= [HW] IO7 for Marvel-based Alpha systems
2498 See Documentation/admin-guide/nfs/nfsroot.rst.
2534 Enables support for pseudo-NMIs in the kernel. This
2553 [Deprecated - use cpusets instead]
2554 Format: [flag-list,]<cpu-list>
2560 Disable the tick when a single task runs as well as
2579 advised to use cpusets instead to disable scheduler load
2581 It offers a much more flexible interface where CPUs can
2584 You can move a process onto or off an "isolated" CPU via
2587 "number of CPUs in system - 1".
2613 The format of <cpu-list> is described above.
2617 ivrs_ioapic [HW,X86-64]
2618 Provide an override to the IOAPIC-ID<->DEVICE-ID
2620 By default, PCI segment is 0, and can be omitted.
2622 For example, to map IOAPIC-ID decimal 10 to
2628 * To map IOAPIC-ID decimal 10 to PCI device 00:14.0
2631 * To map IOAPIC-ID decimal 10 to PCI segment 0x1 and
2635 ivrs_hpet [HW,X86-64]
2636 Provide an override to the HPET-ID<->DEVICE-ID
2638 By default, PCI segment is 0, and can be omitted.
2640 For example, to map HPET-ID decimal 10 to
2646 * To map HPET-ID decimal 0 to PCI device 00:14.0
2649 * To map HPET-ID decimal 10 to PCI segment 0x1 and
2653 ivrs_acpihid [HW,X86-64]
2654 Provide an override to the ACPI-HID:UID<->DEVICE-ID
2656 By default, PCI segment is 0, and can be omitted.
2658 For example, to map UART-HID:UID AMD0020:0 to
2664 * To map UART-HID:UID AMD0020:0 to PCI segment is 0,
2667 * To map UART-HID:UID AMD0020:0 to PCI segment 0x1 and
2691 the kernel for non-movable allocations. The requested
2701 subsystem. Note that allocations like PTEs-from-HighMem
2709 for non-movable allocations and remaining memory is used
2720 gdb or control-c on the dbgp connection. When
2721 not using this parameter you use sysrq-g to break into
2726 or a supported polling keyboard driver (non-usb).
2736 characters and can work in polling mode, you can use
2753 Configure the RouterBoard 532 series on-chip
2756 kmemleak= [KNL,EARLY] Boot-time kmemleak enable/disable
2762 kprobe_event=[probe-list]
2764 The probe-list is a semicolon delimited list of probe
2783 default value can be overridden via
2794 execution by eliminating the write-protection faults
2817 when KVM is loaded, and disable virtualization when KVM
2820 If disabled, KVM will dynamically enable and disable
2821 virtualization on-demand when creating and destroying
2829 is that doing so may interfere with using out-of-tree
2856 back to huge pages. If the value is a non-zero N, KVM will
2861 kvm-amd.nested= [KVM,AMD] Control nested virtualization feature in
2864 kvm-amd.npt= [KVM,AMD] Control KVM's use of Nested Page Tables,
2865 a.k.a. Two-Dimensional Page Tables. Default is 1
2866 (enabled). Disable by KVM if hardware lacks support
2869 kvm-arm.mode=
2873 none: Forcefully disable KVM.
2875 nvhe: Standard nVHE-based mode, without support for
2882 nested: VHE-based mode with support for nested
2887 mode to "protected" will disable kexec and hibernation
2890 command-line.
2894 kvm-arm.vgic_v3_group0_trap=
2895 [KVM,ARM,EARLY] Trap guest accesses to GICv3 group-0
2898 kvm-arm.vgic_v3_group1_trap=
2899 [KVM,ARM,EARLY] Trap guest accesses to GICv3 group-1
2902 kvm-arm.vgic_v3_common_trap=
2906 kvm-arm.vgic_v4_enable=
2910 kvm-arm.wfe_trap_policy=
2919 kvm-arm.wfi_trap_policy=
2936 kvm-intel.ept= [KVM,Intel] Control KVM's use of Extended Page Tables,
2937 a.k.a. Two-Dimensional Page Tables. Default is 1
2938 (enabled). Disable by KVM if hardware lacks support
2941 kvm-intel.emulate_invalid_guest_state=
2943 state. Ignored if kvm-intel.enable_unrestricted_guest=1,
2949 kvm-intel.flexpriority=
2951 (TPR shadow). Default is 1 (enabled). Disable by KVM if
2954 kvm-intel.nested=
2958 kvm-intel.unrestricted_guest=
2961 is 1 (enabled). Disable by KVM if EPT is disabled or
2964 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault
2965 CVE-2018-3620.
2971 VMEXIT and VMENTER can leak host memory.
2976 kvm-intel.vpid= [KVM,Intel] Control KVM's use of Virtual Processor
2978 (enabled). Disable by KVM if hardware lacks support
2985 internal buffers which can forward information to a
2989 forwarded data can be used in a cache side channel
2996 on - enable the interface for the mitigation
3061 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst
3067 lapic [X86-32,APIC,EARLY] Enable the local APIC even if BIOS
3071 value for LAPIC timer one-shot implementation. Default
3079 libata.dma=0 Disable all PATA and SATA DMA
3094 libata.force= [LIBATA] Force configurations. The format is a comma-
3106 first fan-out link behind PMP device. It does not
3113 The following configurations can be forced.
3120 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
3127 * rstonce: only attempt one reset during hot-unplug
3130 * [no]dbdelay: Enable or disable the extra 200ms delay
3136 * [no]ncqtrim: Enable or disable queued DSM TRIM.
3138 * [no]ncqati: Enable or disable NCQ trim on ATI chipset.
3140 * [no]trim: Enable or disable (unqueued) TRIM.
3153 * [no]dmalog: Enable or disable the use of the
3156 * [no]iddevlog: Enable or disable access to the
3159 * [no]logdir: Enable or disable access to the general
3170 * external: Mark port as external (hotplug-capable).
3172 * [no]lpm: Enable or disable link power management.
3177 * [no]fua: Disable or enable FUA (Force Unit Access)
3182 * disable: Disable this device.
3224 Specify the number of self-propagating call_rcu()
3232 occasional long-duration lock hold time. Defaults
3233 to 100 milliseconds. Select 0 to disable.
3238 (MAX_NESTED_LOCKS). Specify zero to disable.
3243 Set the number of locking read-acquisition kthreads.
3248 Set the number of locking write-acquisition kthreads.
3251 Set time (s) after boot for CPU-hotplug testing.
3254 Set time (s) between CPU-hotplug operations, or
3255 zero to disable CPU-hotplug testing.
3258 Do periodic testing of real-time lock priority
3259 boosting. Select 0 to disable, 1 to boost
3263 non-real-time spinlocks, due to their disabling
3264 of preemption. Note that non-realtime mutexes
3265 disable boosting.
3277 Set task-shuffle interval (jiffies). Shuffling
3278 tasks allows some CPUs to go into dyntick-idle
3283 is useful for hands-off automated testing.
3302 Run the write-side locktorture kthreads at
3303 sched_set_fifo() real-time priority.
3310 console loglevel will be printed to the console. It can
3321 7 (KERN_DEBUG) debug-level messages
3333 logo.nologo [FB] Disables display of the built-in Linux logo.
3341 lp=auto printer driver. 'lp=reset' (which can be
3351 an IEEE 1284-compliant printer is attached; if
3357 time-consuming boot-time autodetection (up to 250 ms per
3377 Example: machtype=lemote-yeeloong-2f-7inch
3382 bootup you can bring up the other plugged cpu by executing
3389 (loop.max_loop) unconditionally pre-created at init time. The default
3392 devices can be requested on-demand with the
3393 /dev/loop-control interface.
3395 mce= [X86-{32,64}]
3400 disable machine check
3403 disable CMCI(Corrected Machine Check Interrupt) that
3419 disable features for corrected errors, e.g.
3432 do not opt-in to Local MCE delivery. Use legacy method
3446 disable boot machine check logging.
3450 checks. 0 to disable.
3453 don't overwrite the bios-set CMCI threshold. This boot
3462 force-enable recoverable machine check code paths
3468 See Documentation/admin-guide/md.rst.
3475 Control mitigation for the Micro-architectural Data
3479 internal buffers which can forward information to a
3483 forwarded data can be used in a cache side channel
3490 full - Enable MDS mitigation on vulnerable CPUs
3491 full,nosmt - Enable MDS mitigation and disable
3493 off - Unconditionally disable MDS mitigation
3495 On TAA-affected machines, mds=off can be prevented by
3497 mitigated with the same mechanism so in order to disable
3504 For details see: Documentation/admin-guide/hw-vuln/mds.rst
3519 [ARC,MICROBLAZE] - the limit applies only to low memory,
3522 [ARM64] - only limits memory covered by the linear
3535 [ARM,MIPS,EARLY] - override the memory layout
3539 Multiple different regions can be specified with
3542 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
3549 per-device physically contiguous DMA buffers.
3557 See Documentation/admin-guide/mm/memory-hotplug.rst.
3561 Such memmap=exactmap lines can be constructed based on
3570 Multiple different regions can be specified,
3582 Example: Exclude memory from 0x18690000-0x1869ffff
3596 memmap=<size>%<offset>-<oldtype>+<newtype>
3598 from <oldtype> to <newtype>. If "-<oldtype>" is left
3613 repeatable BIOS-originated corruption always
3614 affects the same memory, you can use memmap=
3642 The state of the flag can be read in
3649 default : 0 <disable>
3657 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control
3663 Refer to Documentation/virt/kvm/x86/amd-memory-encryption.rst
3664 for details on when memory encryption can be activated.
3667 s2idle - Suspend-To-Idle
3668 shallow - Power-On Suspend or equivalent (if supported)
3669 deep - Suspend-To-RAM or equivalent (if supported)
3670 See Documentation/admin-guide/pm/sleep-states.rst.
3672 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
3675 problem by letting the user disable the workaround.
3681 Enable or disable the microcode minimal revision
3688 0 - The attached screen is the 3.5" TFT
3689 1 - The attached screen is the 7" TFT
3690 2 - The VGA Shield is attached (1024x768)
3694 b - Enable backlight. The TFT backlight pin will be
3698 c - Enable the s3c camera interface.
3699 t - Reserved for enabling touchscreen support. The
3701 kernel as of 2.6.30, a preliminary port can be found
3703 https://repo.or.cz/w/linux-2.6/mini2440.git
3708 arch-independent options, each of which is an
3709 aggregation of existing arch-specific options.
3715 Disable all optional CPU mitigations. This
3739 ssbd=force-off [ARM64]
3752 have other ways of avoiding SMT-based attacks.
3787 full - Enable mitigation on vulnerable CPUs
3789 full,nosmt - Enable mitigation and disable SMT on
3792 off - Unconditionally disable mitigation
3795 mmio_stale_data=off can be prevented by an active
3798 disable this mitigation, you need to specify
3805 Documentation/admin-guide/hw-vuln/processor_mmio_stale_data.rst
3810 probe on this module. Otherwise, enable/disable
3816 by default. To enable/disable async probing for a
3835 module_blacklist= [KNL] Do not load a comma-separated list of
3860 movable_node [KNL,EARLY] Boot-time switch to make hotplugable memory
3867 Format: <name>,<region-number>,<size>,<offset>
3870 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
3909 something different and driver-specific.
3920 0 to disable accounting
3949 [NFS] enable 64-bit inode numbers.
3950 If zero, the NFS client will fake up a 32-bit inode
3952 of returning the full 64-bit number.
3953 The default is to return 64-bit inode numbers.
3970 that the client can send to the NFSv4.1 server.
3980 disabling idmapping, which can make migration from
3988 [NFS4] Specify an additional fixed unique ident-
3989 ification string that NFSv4 clients can insert into
4019 driver. A non-zero value sets the minimum interval
4024 server-to-server copies for which this server is
4036 server-to-server copy, knfsd temporarily mounts
4043 See Documentation/admin-guide/nfs/nfsroot.rst.
4045 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
4046 See Documentation/admin-guide/nfs/nfsroot.rst.
4049 See Documentation/admin-guide/nfs/nfsroot.rst.
4053 NMI stack-backtrace request.
4062 0 - turn hardlockup detector in nmi_watchdog off
4063 1 - turn hardlockup detector in nmi_watchdog on
4064 rNNN - configure the watchdog with raw perf event 0xNNN
4069 To disable both hard and soft lockup detectors,
4074 These settings can be accessed at runtime via
4077 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
4081 no4lvl [RISCV,EARLY] Disable 4-level and 5-level paging modes.
4082 Forces kernel to use 3-level paging instead.
4084 no5lvl [X86-64,RISCV,EARLY] Disable 5-level paging mode. Forces
4085 kernel to use 4-level paging instead.
4094 noautogroup Disable scheduler automatic task group creation.
4100 Disable suspending of consoles during suspend and
4102 messages can reach various consoles while the rest
4114 [KNL,EARLY] Disable object debugging
4116 nodsp [SH] Disable hardware DSP at boot time.
4118 noefi [EFI,EARLY] Disable EFI runtime services support.
4120 no_entry_flush [PPC,EARLY] Don't flush the L1-D cache when entering the kernel.
4122 noexec32 [X86-64]
4123 This affects only 32-bit executables.
4124 noexec32=on: enable non-executable mappings (default)
4126 noexec32=off: disable non-executable mappings
4133 nofpu [MIPS,SH] Disable hardware FPU at boot time.
4137 nofxsr [BUGS=X86-32] Disables x86 floating point extended
4139 legacy floating-point registers on task switch.
4151 difficult since unequal pointers can no longer be
4152 compared. However, if this command-line option is
4158 nohibernate [HIBERNATION] Disable hibernation and resume.
4171 nohugeiomap [KNL,X86,PPC,ARM64,EARLY] Disable kernel huge I/O mappings.
4173 nohugevmalloc [KNL,X86,PPC,ARM64,EARLY] Disable kernel huge vmalloc mappings.
4175 nohz= [KNL] Boottime enable/disable dynamic ticks
4195 nointremap [X86-64,Intel-IOMMU,EARLY] Do not enable interrupt
4197 [Deprecated - use intremap=off]
4199 noinvpcid [X86,EARLY] Disable the INVPCID cpu feature.
4203 noirqdebug [X86-32] Disables the code which attempts to detect and
4204 disable unhandled interrupt sources.
4213 no-kvmapf [X86,KVM,EARLY] Disable paravirtualized asynchronous page
4216 no-kvmclock [X86,KVM,EARLY] Disable paravirtualized KVM clock driver
4218 nolapic [X86-32,APIC,EARLY] Do not enable or use the local APIC.
4220 nolapic_timer [X86-32,APIC,EARLY] Do not use the local APIC timer.
4222 nomce [X86-32] Disable Machine Check Exception
4224 nomfgpt [X86-32] Disable Multi-Function General Purpose
4227 nomodeset Disable kernel modesetting. Most systems' firmware
4230 not load if they could possibly displace the pre-
4233 perform display-mode changes or accelerated rendering.
4237 nomodule Disable module load
4239 nonmi_ipi [X86] Disable using NMI IPIs during panic/reboot to
4243 nopat [X86,EARLY] Disable PAT (page attribute table extension of
4246 nopcid [X86-64,EARLY] Disable the PCID cpu feature.
4248 nopku [X86] Disable Memory Protection Keys CPU feature found
4251 nopti [X86-64,EARLY]
4267 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
4273 no-scroll [VGA] Disables scrollback.
4274 This is required for the Braillex ib80-piezo Braille
4277 nosgx [X86-64,SGX,EARLY] Disables Intel SGX kernel support.
4280 Disable SMAP (Supervisor Mode Access Prevention)
4284 Disable SMEP (Supervisor Mode Execution Prevention)
4288 and disable the IO APIC. legacy for "maxcpus=0".
4290 nosmt [KNL,MIPS,PPC,EARLY] Disable symmetric multithreading (SMT).
4293 [KNL,X86,PPC,S390] Disable symmetric multithreading (SMT).
4294 nosmt=force: Force disable SMT, cannot be undone
4297 nosoftlockup [KNL] Disable the soft-lockup detector.
4300 [HW,EARLY] Disable all mitigations for the Speculative
4303 nospectre_bhb [ARM64,EARLY] Disable all mitigations for Spectre-BHB (branch
4307 nospectre_v1 [X86,PPC,EARLY] Disable mitigations for Spectre Variant 1
4311 nospectre_v2 [X86,PPC_E500,ARM64,EARLY] Disable all mitigations
4316 no-steal-acc [X86,PV_OPS,ARM64,PPC/PSERIES,RISCV,LOONGARCH,EARLY]
4317 Disable paravirtualized steal time accounting. steal time
4323 timer IRQ sources, i.e., the IO-APIC timer. This can
4328 [PPC,EARLY] Don't flush the L1-D cache after accessing user data.
4331 Disable device dump. Device dump allows drivers to
4332 append dump data to vmcore so you can collect driver
4333 specified debug info. Drivers can append the data
4336 device dump can help save memory but the driver debug
4341 no-vmw-sched-clock
4342 [X86,PV_OPS,EARLY] Disable paravirtualized VMware
4345 nowatchdog [KNL] Disable both lockup detectors, i.e.
4346 soft-lockup and NMI watchdog (hard-lockup).
4350 nox2apic [X86-64,APIC,EARLY] Do not enable x2APIC mode.
4358 enabling legacy floating-point and sse state.
4379 runtime you can physically add extra cpu until it reaches
4380 n. So during boot up some boot time memory for per-cpu
4381 variables need be pre-allocated for later physical cpu
4387 Disable NUMA, Only set up a single NUMA node
4408 soft-reserved memory partitioning.
4410 numa_balancing= [KNL,ARM64,PPC,RISCV,S390,X86] Enable or disable automatic
4412 Allowed values are enable and disable
4415 'node', 'default' can be specified
4416 This can be set from sysctl after boot.
4417 See Documentation/admin-guide/sysctl/vm.rst for details.
4420 See Documentation/core-api/debugging-via-ohci1394.rst for more
4435 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
4439 boundary - index of last SLC block on Flex-OneNAND.
4441 lock - Configure if Flex-OneNAND boundary should be locked.
4454 should randomize its free lists. This parameter can be
4455 used to enable/disable page randomization. The state of
4456 the flag can be read from sysfs at:
4460 page_owner= [KNL,EARLY] Boot-time page_owner enabling option.
4463 we can turn it on.
4466 page_poison= [KNL,EARLY] Boot-time parameter changing the state of
4490 The optional switch "nousertaint" can be utilized to
4494 See Documentation/admin-guide/tainted-kernels.rst for
4495 extra details on the taint flags that users can pick
4502 User can chose combination of the following bits:
4528 possible conflicts). You can specify the base
4547 Set to non-zero to probe primary and secondary ISA
4553 Set to non-zero if a chip is present that snoops speed
4601 Set to non-zero to probe tertiary and further ISA
4610 of individual ports can be disabled by setting the
4617 Set to non-zero to probe QDI controllers. By default
4622 Set to non-zero to probe Winbond controllers. Use
4659 addresses can be specified after the base
4669 bios [X86-32] force use of PCI BIOS, don't access
4671 has a non-standard PCI host bridge.
4672 nobios [X86-32] disallow use of PCI BIOS, only direct
4678 data in IO port 0xCFC, both 32-bit).
4680 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
4681 the function, IO port 0xCFA, also 8-bit, sets
4683 through ports 0xC000-0xCFFF).
4687 enabled, this kernel boot option can be used to
4688 disable the use of PCIE advanced error reporting.
4689 nodomains [PCI] Disable support for multiple PCI
4690 root domains (aka PCI segments, in ACPI-speak).
4691 nommconf [X86] Disable use of MMCONFIG for PCI
4697 enabled, this kernel boot option can be used to
4698 disable the use of MSI interrupts system-wide.
4699 noioapicquirk [APIC] Disable all boot interrupt quirks.
4703 primary IO-APIC for bridges that cannot disable
4706 noioapicreroute [APIC] Disable workaround that uses the
4710 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
4728 assigned automatically to PCI devices. You can
4734 F0000h-100000h range.
4735 lastbus=N [X86] Scan all buses thru bus #N. Can be
4739 assign-busses [X86] Always assign all PCI bus
4760 <linux-pci@vger.kernel.org>.
4764 a bug to <linux-pci@vger.kernel.org>.
4769 skip_isa_align [X86] do not align io start addr, so can
4776 bfsort Sort PCI devices into breadth-first order.
4779 nobfsort Don't sort PCI devices into breadth-first order.
4780 pcie_bus_tune_off Disable PCIe MPS (Max Payload Size)
4781 tuning and use the BIOS-configured MPS defaults.
4788 or bus can support) for best performance.
4791 configuration allows peer-to-peer DMA between
4794 that hot-added devices will work.
4809 A PCI-PCI bridge can be specified if resource
4815 for 4096-byte alignment.
4816 ecrc= Enable/disable PCIe ECRC (transaction layer
4817 end-to-end CRC checking). Only effective if
4840 realloc= Enable/disable reallocating PCI bridge resources
4848 noats [PCIE, Intel-IOMMU, AMD-IOMMU]
4855 can resize a BAR to allow access to all VRAM.
4879 bit-0 : ACS Source Validation
4880 bit-1 : ACS Translation Blocking
4881 bit-2 : ACS P2P Request Redirect
4882 bit-3 : ACS P2P Completion Redirect
4883 bit-4 : ACS Upstream Forwarding
4884 bit-5 : ACS P2P Egress Control
4885 bit-6 : ACS Direct Translated P2P
4886 Each bit can be marked as:
4893 ACS to enable P2P Request Redirect, disable
4895 Validation unchanged from whatever power-up
4905 is enabled, this kernel boot option can be used
4906 to disable PCIe TLP Processing Hints support
4907 system-wide.
4921 dpc-native Use native PCIe service for DPC only. May
4923 compat Disable native PCIe services (PME, AER, DPC, PCIe
4927 off Disable power management of all PCIe ports
4938 Keep all power-domains already enabled by bootloader on,
4943 pdcchassis= [PARISC,HW] Disable/Enable PDC Chassis Status codes at
4956 pirq= [SMP,APIC] Manual mp-table setup
4957 See Documentation/arch/x86/i386/IO-APIC.rst.
4961 See also Documentation/admin-guide/parport.rst.
4980 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
4989 { on | off | curr | res | no-curr | no-res }
5030 Disable Hardware Transactional Memory
5034 none - Limited to cond_resched() calls
5035 voluntary - Limited to cond_resched() and might_sleep() calls
5036 full - Any section that isn't explicitly preempt disabled
5037 can be preempted anytime. Tasks will also yield
5040 lazy - Scheduler controlled. Similar to full but instead
5046 print-fatal-signals=
5052 coredump - etc.
5055 you might want to try "ulimit -i unlimited".
5062 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
5066 Disable console loglevel raise on oops, panic
5067 or lockdep-detected issues (only if lock debug is on).
5075 Allows storing messages from non-panic CPUs into
5077 flushed to consoles by the panic-CPU on
5078 a best-effort basis.
5079 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
5084 on - unlimited logging to /dev/kmsg from userspace
5085 off - logging to /dev/kmsg disabled
5086 ratelimit - ratelimit the logging
5090 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
5096 restrict that. Can be one of:
5097 - 'always': traditional behavior always allows mem overrides.
5098 - 'ptrace': only allow mem overrides for active ptracers.
5099 - 'never': never allow mem overrides.
5103 Limit processor to maximum C-state
5107 Ignore the _CST method to determine C-states,
5114 Param: "schedule" - profile schedule points.
5115 Param: "kvm" - profile VM exits.
5116 Param: <number> - step/bucket size as a power of 2 for
5128 psi= [KNL] Enable or disable pressure stall information
5147 pti= [X86-64] Control Page Table Isolation of user and
5152 on - unconditionally enable
5153 off - unconditionally disable
5154 auto - kernel detects whether your CPU model is
5160 [KNL] Number of legacy pty's. Overwrites compiled-in
5163 quiet [KNL,EARLY] Disable most log messages
5168 Disable RADIX GTSE feature and use hcall for TLB
5172 See Documentation/admin-guide/md.rst.
5175 See Documentation/admin-guide/blockdev/ramdisk.rst.
5180 [KNL,EARLY] Disable trusting the use of the CPU's
5185 [KNL,EARLY] Disable trusting the use of the a seed
5190 [KNL,EARLY] Enable or disable kernel stack offset
5194 cross-syscall address exposures. This is only
5197 Format: <bool> (1/Y/y=enable, 0/N/n=disable)
5200 ras=option[,option,...] [KNL] RAS-specific options
5203 Disable the Correctable Errors Collector,
5206 rcu_nocbs[=cpu-list]
5211 enable the no-callback CPU mode, which prevents
5216 "p" for RCU-preempt, "s" for RCU-sched, and "g"
5219 the offloaded CPUs, which can be useful for HPC
5220 and real-time workloads. It can also improve
5224 list of CPUs is set to no-callback mode from boot.
5228 no-callback mode from boot but the mode may be
5239 This improves the real-time response for the
5250 Do only a one-line RCU CPU stall warning when
5251 there is an ongoing too-long CSD-lock wait.
5255 throttled so that userspace tests can safely
5257 If triggered before the RCU grace-period machinery
5267 RCU grace-period cleanup.
5271 RCU grace-period initialization.
5275 RCU grace-period pre-initialization, that is,
5276 the propagation of recent CPU-hotplug changes up
5280 Set delay from grace-period initialization to
5293 soliciting quiescent-state help from
5305 Set the SCHED_FIFO priority of the RCU per-CPU
5308 and for the RCU grace-period kthreads (rcu_bh,
5310 set, valid values are 1-99 and the default is 1
5311 (the least-favored priority). Otherwise, when
5312 RCU_BOOST is not set, valid values are 0-99 and
5313 the default is zero (non-realtime operation).
5318 On callback-offloaded (rcu_nocbs) CPUs,
5321 use of the ->nocb_bypass list. However, in the
5322 common non-flooded case, RCU queues directly to
5323 the main ->cblist in order to avoid the extra
5324 overhead of the ->nocb_bypass list and its lock.
5326 a single jiffy, RCU pre-queues the callbacks into
5327 the ->nocb_bypass queue. The definition of "too
5331 On callback-offloaded (rcu_nocbs) CPUs, avoid
5344 batch limiting is re-enabled.
5348 RCU's force-quiescent-state scan will aggressively
5353 disable more aggressive help enlistment.
5356 Set the page-cache refill delay (in milliseconds)
5357 in response to low-memory conditions. The range
5361 Set the shift-right count to use to compute
5362 the callback-invocation batch limit bl from
5369 Please note that this callback-invocation batch
5370 limit applies only to non-offloaded callback
5376 Disable autobalancing of the rcu_node combining
5379 cache-to-cache transfer latencies.
5385 and for NUMA systems with large remote-access
5401 the wakeup overhead on the global grace-period
5403 each group's NOCB grace-period kthread.
5406 Cause the grace-period kthread to get an extra
5408 it should at force-quiescent-state time.
5421 this specifies an rcu_read_unlock()-time delay
5425 of RCU-protected pointers after the relevant
5435 per-CPU rcuc kthreads. Defaults to a non-zero
5439 But note that CONFIG_PREEMPT_RT=y kernels disable
5464 grace-period primitives such as call_rcu().
5475 grace-period primitives.
5478 Set test-start holdoff period. The purpose of
5496 Test the double-argument variant of kfree_rcu().
5498 rcuscale.kfree_rcu_test_single, both the single-
5499 and double-argument variants are tested.
5502 Test the single-argument variant of kfree_rcu().
5504 rcuscale.kfree_rcu_test_double, both the single-
5505 and double-argument variants are tested.
5519 does not affect the data-collection interval,
5524 Set number of RCU readers. The value -1 selects
5526 "n" less than -1 selects N-n+1, where N is again
5527 the number of CPUs. For example, -2 selects N
5528 (the number of CPUs), -3 selects N+1, and so on.
5529 A value of "n" less than or equal to -N selects
5542 complete. This is useful for hands-off automated
5549 Write-side holdoff between grace periods,
5554 Additional write-side holdoff between grace
5572 for RCU grace-period forward-progress testing
5579 Specify the fraction of a CPU-stall-warning
5580 period to do tight-loop forward-progress testing.
5584 forward-progress tests.
5588 need_resched() during tight-loop forward-progress
5592 Use conditional/asynchronous update-side
5593 normal-grace-period primitives, if available.
5596 Use conditional/asynchronous update-side
5597 expedited-grace-period primitives, if available.
5600 Use conditional/asynchronous update-side
5601 normal-grace-period primitives that also take
5606 Use conditional/asynchronous update-side
5607 expedited-grace-period primitives that also take
5631 Use expedited update-side primitives, if available.
5634 Use normal (non-expedited) asynchronous
5635 update-side primitives, if available.
5638 Use polled update-side normal-grace-period
5642 Use polled update-side expedited-grace-period
5646 Use polled update-side normal-grace-period
5651 Use polled update-side expedited-grace-period
5675 Use normal (non-expedited) synchronous
5676 update-side primitives, if available. If all
5680 they are all non-zero.
5683 Enable grace-period wrap lag testing. Setting
5688 Set the value for grace-period wrap lag during
5703 the grace-period wrap lag will be set to the
5713 Leak an RCU-protected pointer out of the reader.
5714 This can of course result in splats, and is
5728 Set number of RCU callback-offload togglers.
5733 callback-offload toggling attempts.
5736 Set number of RCU readers. The value -1 selects
5737 N-1, where N is the number of CPUs. A value
5738 "n" less than -1 selects N-n-2, where N is again
5739 the number of CPUs. For example, -2 selects N
5740 (the number of CPUs), -3 selects N+1, and so on.
5743 Enable debug-object double-call_rcu() testing.
5746 Set time (s) after boot for CPU-hotplug testing.
5749 Set time (jiffies) between CPU-hotplug operations,
5750 or zero to disable CPU-hotplug testing.
5754 by a high-priority FIFO real-time task. Set to
5755 zero (the default) to disable. The CPUs to
5763 second) between preemptions by a high-priority
5764 FIFO real-time task. This delay is mediated
5769 The number of times in a given read-then-exit
5770 episode that a set of read-then-exit kthreads
5775 read-then-exit testing episodes.
5780 are entered from low-order bit up, and are
5782 0x1 bit is normal readers, 0x2 NMI-safe readers,
5783 and 0x4 light-weight readers.
5786 Set task-shuffle interval (s). Shuffling tasks
5787 allows some CPUs to go into dyntick-idle mode
5792 is useful for hands-off automated testing.
5796 warnings, zero to disable.
5801 any other stall-related activity. Note that
5807 in scheduling-while-atomic splats.
5816 Disable interrupts while stalling if set, but only
5826 grace-period kthread to test RCU CPU stall
5827 warnings, zero to disable. If both stall_cpu
5850 of RCU priority-boost testing. Defaults to zero,
5857 Test RCU's dyntick-idle handling. See also the
5912 Use expedited grace-period primitives, for
5915 but can increase CPU utilization, degrade
5916 real-time latency, and degrade energy efficiency.
5920 Use only normal grace-period primitives,
5923 real-time latency, CPU utilization, and
5924 energy efficiency, but can expose users to
5925 increased grace-period latency. This parameter
5932 only normal grace-period primitives. No effect
5938 post-boot attempt at an expedited RCU grace
5939 period to instead use normal non-expedited
5940 grace-period processing.
5948 set to the default value of -1.
5951 Set the minimum number of callback-queuing-time
5952 lock-contention events per jiffy required to
5953 cause the RCU Tasks flavors to switch to per-CPU
5956 the default value of -1.
5961 of -1 allows this to be automatically (and
5969 number avoids disturbing real-time workloads,
5974 cancel laziness on that CPU. Use -1 to disable
5984 only printed prior to the stall-warning message
5985 for a given grace period. Disable with a value
6003 warning messages. Disable with a value less
6012 of zero will disable batching. Batching is
6020 disable batching. Batching is always disabled
6032 force - Override the decision by the kernel to hide the
6071 There are some built-in platform specific "quirks"
6072 - you may see: "reboot: <name> series board detected.
6076 built-in quirk table, and use the generic default
6085 Don't stop other CPUs on reboot. This can make reboot
6109 Set test-start holdoff period. The purpose of
6124 but the default has already reduced the per-pass
6129 Set number of readers. The default value of -1
6138 Set the read-side critical-section duration,
6142 Specify the read-protection implementation to test.
6168 See Documentation/admin-guide/cgroup-v1/cpusets.rst.
6179 other subsystems can use to access it. This is typically
6184 or if booting a different kernel. It can also fail if KASLR
6198 reservetop= [X86-32,EARLY]
6215 See Documentation/power/swsusp-and-swap-files.rst
6231 AMD-based UNRET and IBPB mitigations alone do not stop
6233 sibling threads. For that reason, STIBP is used on pro-
6237 off - no mitigation
6238 auto - automatically select a migitation
6239 auto,nosmt - automatically select a mitigation,
6243 ibpb - On AMD, mitigate short speculation
6248 ibpb,nosmt - Like "ibpb" above but will disable SMT
6252 unret - Force enable untrained return thunks,
6253 only effective on AMD f15h-f17h based
6255 unret,nosmt - Like unret, but will disable SMT when STIBP
6276 ring3mwait=disable
6277 [KNL] Disable ring 3 MONITOR/MWAIT feature on supported
6287 ro [KNL] Mount root device read-only on boot
6290 on Mark read-only kernel memory as read-only (default).
6291 off Leave read-only kernel memory writable for debugging.
6292 full Mark read-only kernel memory and aliases as read-only
6299 debug-uart get routed to the D+ and D- pins of the usb
6305 block/early-lookup.c for details.
6306 Alternatively this can be "ram" for the legacy initial
6330 rt_group_sched= [KNL] Enable or disable SCHED_RR/FIFO group scheduling
6335 rw [KNL] Mount root device read-write on boot
6351 The default is 1 meaning that one can concurrently use
6354 once. With a value of 2 one can map all of memory twice
6364 schedstats= [KNL,X86] Enable or disable scheduled statistics.
6365 Allowed values are enable and disable. This feature
6389 to 10 seconds for built-in smp_call_function()
6396 that requesting even small non-zero numbers of
6397 seconds can result in RCU CPU stall warnings,
6403 The default of -1 specifies a number of kthreads
6408 test before initiating CPU-hotplug operations.
6412 CPU-hotplug operations. Specifying zero (which
6413 is the default) disables CPU-hotplug operations.
6419 Non-zero values are useful for automated tests.
6432 preempt_disable() to disable CPU hotplug
6442 "wait" parameter. A value of -1 selects the
6443 default if all other weights are -1. However,
6445 value of -1 will instead select a weight of zero.
6450 non-zero "wait" parameter. See weight_single.
6457 this weighting can place serious IPI load
6463 non-zero "wait" parameter. See weight_single
6475 non-zero "wait" parameter. See weight_single
6486 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
6487 1 -- enable.
6495 selinux= [SELINUX] Disable or enable SELinux at boot time.
6498 0 -- disable.
6499 1 -- enable.
6502 serialnumber [BUGS=X86-32]
6504 sev=option[,option...] [X86-64]
6510 Do not enable SEV-SNP (applies to host/hypervisor
6513 SEV-SNP guests.
6521 to set it to "all" by meaning -- no limit here.
6530 slab_debug can create guard zones around objects and
6565 Disable merging of slabs with similar size. May be
6569 layout control by attackers can usually be
6573 unchanged). Debug options disable merging on their
6602 using the csdlock_debug- kernel parameter.
6607 system. By default, let CSD-lock acquisition
6609 for this value provides a 5-minute timeout.
6611 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
6612 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
6613 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
6614 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
6615 smsc-ircc2.ircc_irq= [HW] IRQ line
6616 smsc-ircc2.ircc_dma= [HW] DMA channel
6617 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
6627 Default: -1 (no limit)
6630 [KNL] Should the soft-lockup detector generate panics.
6633 A value of 1 instructs the soft-lockup detector
6634 to panic the machine when a soft-lockup occurs. It is
6637 respective build-time switch to that functionality.
6640 [KNL] Should the soft-lockup detector generate
6645 See Documentation/admin-guide/laptops/sonypi.rst
6652 on - (default) Enable the HW or SW mitigation as
6655 vmexit - On systems which don't have the HW mitigation
6658 protected from VM-originated BHI attacks, but
6660 off - Disable the mitigation.
6667 on - unconditionally enable, implies
6669 off - unconditionally disable, implies
6671 auto - kernel detects whether your CPU model is
6685 Selecting 'off' will disable both the kernel and
6688 Specific mitigations can also be selected manually:
6690 retpoline - replace indirect branches
6691 retpoline,generic - Retpolines
6692 retpoline,lfence - LFENCE; indirect branch
6693 retpoline,amd - alias for retpoline,lfence
6694 eibrs - Enhanced/Auto IBRS
6695 eibrs,retpoline - Enhanced/Auto IBRS + Retpolines
6696 eibrs,lfence - Enhanced/Auto IBRS + LFENCE
6697 ibrs - use IBRS to protect kernel
6707 on - Unconditionally enable mitigations. Is
6710 off - Unconditionally disable mitigations. Is
6713 prctl - Indirect branch speculation is enabled,
6714 but mitigation can be enabled via prctl
6719 - Like "prctl" above, but only STIBP is
6725 - Same as "prctl" above, but all seccomp
6730 - Like "seccomp" above, but only STIBP is
6735 auto - Kernel selects the mitigation depending on
6746 off - Disable mitigation
6747 microcode - Enable microcode mitigation only
6748 safe-ret - Enable sw-only safe RET mitigation (default)
6749 ibpb - Enable mitigation by issuing IBPB on
6751 ibpb-vmexit - Issue IBPB only on VMEXIT
6752 (cloud-specific mitigation)
6755 [HW,EARLY] Control Speculative Store Bypass (SSB) Disable mitigation
6763 is that such stores are unlikely and that they can
6768 store can be used in a cache side channel attack, for
6777 on - Unconditionally disable Speculative Store Bypass
6778 off - Unconditionally enable Speculative Store Bypass
6779 auto - Kernel detects whether the CPU model contains an
6785 prctl - Control Speculative Store Bypass per thread
6789 seccomp - Same as "prctl" above, but all seccomp threads
6790 will disable SSB unless they explicitly opt out.
6797 on,auto - On Power8 and Power9 insert a store-forwarding
6801 off - No action.
6815 off - not enabled
6817 warn - the kernel will emit rate-limited warnings
6825 fatal - the kernel will send SIGBUS to applications
6830 ratelimit:N -
6850 Certain CPUs are vulnerable to an MDS-like
6851 exploit which can leak bits from the random
6855 microcode. However, the microcode fix can cause
6860 The microcode mitigation can be disabled with
6863 off: Disable mitigation and remove
6870 This kernel-boot parameter defaults to 128,
6871 but takes effect only when the low-order four
6888 instead of the compile-time CONFIG_NR_CPUS.
6892 grace-period sequence counter wrap for the
6893 srcu_data structure's ->srcu_gp_seq_needed field.
6904 expediting. Set to zero to disable automatic
6908 Specifies the number of no-delay instances
6915 Specifies the per-grace-period phase, number of
6916 non-sleeping polls of readers. Beyond this limit,
6922 Specifies number of microseconds of non-sleeping
6923 delay between each non-sleeping poll of readers.
6926 Specifies the number of update-side contention
6931 set for contention-based conversions to occur.
6934 Speculative Store Bypass Disable control
6941 force-on: Unconditionally enable mitigation for
6943 force-off: Unconditionally disable mitigation for
6959 disable the stack depot thereby saving the static memory
6966 stacktrace_filter=[function-list]
6968 will trace at boot up. function-list is a comma-separated
6969 list of functions. This list can be changed at run
6976 Set the STI (builtin display/keyboard on the HP-PARISC
6978 as the initial boot-console.
6990 Enable or disable strict sigaltstack size checks
6992 depends on the supported FPU features. This can
7038 to global on non-NUMA machines)
7044 RPC calls that can be sent from the client to a
7063 <int> -- Number of I/O TLB slabs
7064 <int> -- Second integer after comma. Number of swiotlb
7067 force -- force using of bounce buffers even if they
7069 noforce -- Never use bounce buffers (for debugging)
7084 Ignore sysrq setting - this boot parameter will
7092 cache size. See Documentation/networking/ip-sysctl.rst
7099 Specify "mem" (for Suspend-to-RAM) or "standby" (for
7104 wakeup-capable RTC alarm.
7110 -1: disable all active trip points in all thermal zones
7114 -1: disable all critical trip points in all thermal zones
7118 1: disable ACPI thermal control
7121 -1: disable all passive trip points
7127 <deci-seconds>: poll all this frequency
7131 Format: <size>[KMG],<size>[KMG]:<state>;<size>[KMG]-<size>[KMG]:<state>
7135 Can be used multiple times for multiple anon THP sizes.
7136 See Documentation/admin-guide/mm/transhuge.rst for more
7144 Format: <size>[KMG],<size>[KMG]:<policy>;<size>[KMG]-<size>[KMG]:<policy>
7149 It can be used multiple times for multiple shmem THP sizes.
7150 See Documentation/admin-guide/mm/transhuge.rst for more
7162 Prevent the CPU-hotplug component of torturing
7166 Dump the ftrace buffer at torture-test shutdown,
7167 even if there were no errors. This can be a
7170 with rotating-rust storage.
7175 disables verbose-printk() sleeping.
7178 Duration of each verbose-printk() sleep in jiffies.
7203 https://trustedcomputinggroup.org/resource/pc-client-platform-tpm-profile-ptp-specification/
7217 The tp_printk_stop_on_boot (see below) can also be used
7224 frequency tracepoints such as irq or sched, can cause
7228 When tp_printk (above) is set, it can cause a lot of noise
7241 local - Use the per CPU time stamp counter
7245 global - Event time stamps are synchronize across
7248 counter - Simple counting of events (1, 2, ..)
7252 uptime - Use jiffies as the time stamp.
7253 perf - Use the same clock that perf uses.
7254 mono - Use ktime_get_mono_fast_ns() for time stamps.
7255 mono_raw - Use ktime_get_raw_fast_ns() for time
7257 boot - Use ktime_get_boot_fast_ns() for time stamps.
7261 trace_event=[event-list]
7263 to facilitate early boot debugging. The event-list is a
7264 comma-separated list of trace events to enable. See
7275 trace_instance=[instance-info]
7281 Events can be enabled at the time the instance is created
7295 Flags can be added to the instance to modify its behavior when it is
7300 traceoff - Have the tracing instance tracing disabled after it is created.
7301 traceprintk - Have trace_printk() write into this trace instance
7302 (note, "printk" and "trace_printk" can also be used)
7309 can use that memory:
7317 Alternatively, the memory can be reserved by the reserve_mem option:
7331 it is recommend to disable tracing so that events from a previous boot do not
7339 can force a memory reset on boot which will clear any trace that was stored.
7340 This is just one of many ways that can clear memory. Make sure your system
7348 trace_options=[option-list]
7349 [FTRACE] Enable or disable tracer options at boot.
7350 The option-list is a comma delimited list of options
7351 that can be enabled or disabled just as if you were
7364 trace_trigger=[trigger-list]
7387 disable tracing after the boot is finished, otherwise
7393 [FTRACE] enable this option to disable tracing when a
7394 warning is hit. This turns off "tracing_on". Tracing can
7402 This option can also be set at run time via the sysctl
7408 Can be used to control the default behavior of the system
7410 See Documentation/admin-guide/mm/transhuge.rst
7415 Can be used to control the hugepage allocation policy for
7417 See Documentation/admin-guide/mm/transhuge.rst
7422 Can be used to control the default hugepage allocation policy
7424 See Documentation/admin-guide/mm/transhuge.rst
7432 - "tpm"
7433 - "tee"
7434 - "caam"
7435 - "dcp"
7444 Can be one of:
7445 - "kernel"
7446 - the same value as trusted.source: "tpm" or "tee"
7447 - "default"
7458 trusted.source=dcp and will disable the check if the
7463 tsc= Disable clocksource stability checks for TSC.
7468 high-resolution timer mode on older hardware, and in
7471 Used to run time disable IRQ_TIME_ACCOUNTING on any
7473 can add overhead.
7477 [x86] nowatchdog: disable clocksource watchdog. Used
7489 can be overridden by a later tsc=nowatchdog. A console
7504 on - Enable TSX on the system. Although there are
7507 several previous speculation-related CVEs, and
7511 off - Disable TSX on the system. (Note that this
7519 auto - Disable TSX if X86_BUG_TAA is present,
7524 See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
7530 Similar to Micro-architectural Data Sampling (MDS)
7533 exploit against CPU internal buffers which can forward
7538 data can be used in a cache side channel attack, to
7545 full - Enable TAA mitigation on vulnerable CPUs
7548 full,nosmt - Enable TAA mitigation and disable SMT on
7551 vulnerable to cross-thread TAA attacks.
7552 off - Unconditionally disable TAA mitigation
7554 On MDS-affected machines, tsx_async_abort=off can be
7556 are mitigated with the same mechanism so in order to disable
7566 Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
7572 See also Documentation/input/devices/joystick-parport.rst
7574 udbg-immortal [PPC] When debugging early kernel crashes that
7580 Set number of hash buckets for UDP/UDP-Lite connections
7582 uhci-hcd.ignore_oc=
7584 Some badly-designed motherboards generate lots of
7609 unwind_debug [X86-64,EARLY]
7610 Enable unwinder debug output. This can be
7617 (default -1 = authorized (same as 1),
7623 for newly-detected USB devices (default 2). This
7651 [USB] Specifies timeout for the initial 64-byte
7655 usbcore.nousb [USB] Disable the USB subsystem
7658 [USB] A list of quirk entries to augment the built-in
7661 VendorID:ProductID:Flags. The IDs are 4-digit hex
7663 will change the built-in quirk; setting it if it is
7668 a 255-byte read);
7669 b = USB_QUIRK_RESET_RESUME (device can't resume
7671 c = USB_QUIRK_NO_SET_INTF (device can't handle
7672 Set-Interface requests);
7673 d = USB_QUIRK_CONFIG_INTF_STRINGS (device can't
7676 e = USB_QUIRK_RESET (device can't be reset
7680 bNumInterfaces count, and can't handle
7691 (bInterval-1).
7696 i = USB_QUIRK_DEVICE_QUALIFIER (device can't
7702 k = USB_QUIRK_NO_LPM (device can't handle Link
7729 usb-storage.delay_use=
7736 usb-storage.quirks=
7738 override the built-in unusual_devs list. List
7741 and Product ID values (4-digit hex numbers) and
7743 to a common usb-storage quirk flag as follows:
7785 medium is write-protected).
7794 1 - undefined instruction events
7795 2 - system calls
7796 4 - invalid data aborts
7797 8 - SIGSEGV faults
7798 16 - SIGBUS faults
7805 vdso=0: disable VDSO mapping
7807 vdso32= [X86] Control the 32-bit vDSO
7808 vdso32=1: enable 32-bit VDSO
7809 vdso32=0 or vdso32=2: disable 32-bit VDSO
7819 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
7839 <size> := size (can use standard suffixes
7848 Can be used multiple times for multiple devices.
7850 vga= [BOOT,X86-32] Select a particular video mode
7852 Documentation/admin-guide/svga.rst.
7867 - Disable all of the above options
7870 exact size of <nn>. This can be used to increase
7872 It can also be used to decrease the size and leave more room
7891 vsyscall= [X86-64,EARLY]
7894 code). Most statically-linked binaries and older
7897 targets for exploits that can control RIP.
7918 see vga-softcursor.rst. Default: 2 = underline.
7923 This is a 16-member array composed of values
7924 ranging from 0-255.
7929 This is a 16-member array composed of values
7930 ranging from 0-255.
7935 This is a 16-member array composed of values
7936 ranging from 0-255.
7941 Set system-wide default UTF-8 mode for all tty's.
7942 Default is 1, i.e. UTF-8 mode is enabled for all
7947 Format=<-1|0|1>
7948 Set system-wide default for whether a cursor
7949 is shown on new VTs. Default is -1,
7954 vt.italic= [VT] Default color for italic text; 0-15.
7957 vt.underline= [VT] Default color for underlined text; 0-15.
7961 see Documentation/watchdog/watchdog-parameters.rst
7962 or other driver-specific files in the
7976 Format: <cpu-list>
7981 If CONFIG_WQ_WATCHDOG is configured, workqueue can
7986 it can be updated at runtime by writing to the
7997 Per-cpu work items which run for longer than this
8000 them from noticeably delaying other per-cpu work
8018 Per-cpu workqueues are generally preferred because
8020 locality; unfortunately, per-cpu workqueues tend to
8023 Enabling this makes the per-cpu workqueues which
8034 workqueues. Can be one of "cpu", "smt", "cache",
8037 Documentation/core-api/workqueue.rst.
8039 This can be changed after boot by writing to the
8050 forces round-robin CPU selection to flush out
8058 on - Enable writecombine, use WUC for ioremap_wc()
8059 off - Disable writecombine, use SUC for ioremap_wc()
8061 x2apic_phys [X86-64,APIC,EARLY] Use x2apic physical mode instead of
8065 xen_512gb_limit [KNL,X86-64,XEN]
8075 ide-disks -- unplug primary master IDE devices
8076 aux-ide-disks -- unplug non-primary-master IDE devices
8077 nics -- unplug network devices
8078 all -- unplug all emulated devices (NICs and IDE disks)
8079 unnecessary -- unplugging emulated devices is
8082 never -- do not unplug even if version check succeeds
8096 Select whether to always use non-faulting (safe) MSR
8107 [KNL,X86,XEN,EARLY] Disable the vector callback for Xen
8112 to Xen, for use by other domains. Can be also changed at runtime
8116 xen_timer_slop= [X86-64,XEN,EARLY]
8139 Boolean parameter to disable using fifo event handling
8141 preferred over the 2-level event handling, as it is
8158 xive.store-eoi=off [PPC]
8164 xhci-hcd.quirks [USB,KNL]
8166 host controller quirks. Meaning of each bit can be
8182 can be written using xmon commands.
8184 memory, and other data can't be written using