Lines Matching +full:no +full:- +full:tick +full:- +full:in +full:- +full:suspend
5 force -- enable ACPI if default was off
6 on -- enable ACPI but allow fallback to DT [arm64,riscv64]
7 off -- disable ACPI if default was on
8 noirq -- do not use ACPI for IRQ routing
9 strict -- Be less tolerant of platforms that are not
11 rsdt -- prefer RSDT over (default) XSDT
12 copy_dsdt -- copy DSDT to memory
26 If set to vendor, prefer vendor-specific driver
51 debug output. Bits in debug_layer correspond to a
52 _COMPONENT in an ACPI source file, e.g.,
54 Bits in debug_level correspond to a level in
58 Documentation/firmware-guide/acpi/debug.rst for more information about
74 { strict | lax | no }
77 only). IO ports and memory declared in ACPI might be
78 used by the ACPI subsystem in arbitrary AML code and
86 no: ACPI OperationRegions are not marked as reserved,
87 no further checks are performed.
96 default in APIC mode
100 default in PIC mode
111 by unsupported hardware/firmware features can result in
116 Format: <byte> or <bitmap-list>
119 Disable auto-serialization of AML methods
122 auto-serialization feature.
131 By default, SSDTs contained in the RSDT/XSDT will be
140 Ignore the ACPI-based watchdog interface (WDAT) and let
154 be necessary to carry out a cold reboot _twice_ in a
161 acpi_osi=! # disable all built-in OS vendor
163 acpi_osi=!! # enable all built-in OS vendor
167 'acpi_osi=!' can be used in combination with single or
182 'acpi_osi=' cannot be used in combination with other
184 exist in the ACPI namespace. NOTE that such command can
192 'acpi_osi=!*' can be used in combination with single or
225 For broken nForce2 BIOS resulting in XT-PIC timer.
234 as soon as the kernel's real-mode entry point is called.
250 ACPI NVS memory during suspend/hibernation and resume.
255 behave incorrectly in some ways with respect to system
256 suspend and resume to be ignored (use wisely).
262 add_efi_memmap [EFI; X86] Include EFI memory map in
272 See Documentation/sound/alsa-configuration.rst
279 align_va_addr= [X86-64]
284 CPU-intensive style benchmark, and it can vary highly in
287 32: only for 32-bit processes
288 64: only for 64-bit processes
289 on: enable for both 32- and 64-bit processes
290 off: disable for both 32- and 64-bit processes
300 Allow execve() of 32-bit applications and setting of the
302 subset of the CPUs support 32-bit EL0. When this
303 parameter is present, the set of CPUs supporting 32-bit
305 and hot-unplug operations may be restricted.
307 See Documentation/arch/arm64/asymmetric-32bit.rst for more
310 amd_iommu= [HW,X86-64]
311 Pass parameters to the AMD IOMMU driver in the system.
313 fullflush - Deprecated, equivalent to iommu.strict=1
314 off - do not initialize any AMD IOMMU found in
316 force_isolation - Force device isolation for all
321 force_enable - Force enable the IOMMU on platforms known
324 pgtbl_v1 - Use v1 page table for DMA-API (Default).
325 pgtbl_v2 - Use v2 page table for DMA-API.
326 irtcachedis - Disable Interrupt Remapping Table (IRT) caching.
328 amd_iommu_dump= [HW,X86-64]
334 amd_iommu_intr= [HW,X86-64]
337 legacy - Use legacy interrupt remapping mode.
338 vapic - Use virtual APIC mode, which allows IOMMU
340 This mode requires kvm-amd.avic=1.
349 In this mode autonomous selection is disabled.
363 selects a performance level in this range and appropriate
377 Power management functions (SPARCstation-4/5 + deriv.)
379 Disable APC CPU standby support. SPARCstation-Fox does
380 not play well with APC CPU idle - disable it if you have
387 when initialising the APIC and IO-APIC components.
388 For X86-32, this can also be used to specify an APIC
411 0 -- disable.
412 1 -- enable.
415 arcrimi= [HW,NET] ARCnet - "RIM I" (entirely mem-mapped) cards
457 audit= [KNL] Enable the audit sub-system
459 0 | off - kernel audit is disabled and can not be
461 unset - kernel audit is initialized but disabled and
463 1 | on - kernel audit is initialized and partially
465 messages in RAM until it is fully enabled by the
476 0 - Disable the BAU.
477 1 - Enable the BAU.
478 unset - Disable the BAU.
505 See Documentation/block/cmdline-partition.rst
518 See Documentation/admin-guide/bootconfig.rst
523 bttv.pll= See Documentation/admin-guide/media/bttv.rst
532 cachesize= [BUGS=X86-32] Override level 2 CPU cache size detection.
540 [NET] Specifies amount of time (in seconds) that
551 inclusive. See arch/mips/include/asm/pgtable-bits.h
561 - foo isn't auto-mounted if you mount all cgroups in
563 - foo isn't visible as an individually mountable
565 - if foo is an optional feature then the feature is
571 Specifying "pressure" disables per-cgroup pressure
574 cgroup_no_v1= [KNL] Disable cgroup controllers and named hierarchies in v1
578 the blacklisted controllers remain available in cgroup2.
585 nosocket -- Disable socket memory accounting.
586 nokmem -- Disable kernel memory accounting.
587 nobpf -- Disable BPF memory accounting.
592 0 -- check protection applied by kernel (includes
594 1 -- check protection requested by application.
606 numbers X. Note the Linux-specific bits are not necessarily
607 stable over kernel options, but the vendor-specific
609 X can also be a string as appearing in the flags: line
610 in /proc/cpuinfo which does not have the above
612 in /proc/cpuinfo.
617 being used by the kernel or shown in /proc/cpuinfo.
625 device driver but are enabled in hardware at reset or
627 force such clocks to be always-on nor does it reserve
628 those clocks in any way. This parameter is useful for
631 information, see Documentation/driver-api/clk.rst.
633 clock= [BUGS=X86-32, HW] gettimeofday clocksource override.
650 [X86-32] pit,hpet,tsc;
656 [SPARC64] tick
657 [X86-64] hpet,tsc
663 architected timer so that code using WFE-based polling
675 no replacement if the same CPU is chosen twice.
677 clocksource-wdtest.holdoff= [KNL]
678 Set the time in seconds that the clocksource
683 cma=nn[MG]@[start[MG][-end[MG]]]
694 Sets the size of kernel per-numa memory area for
696 per-numa CMA altogether. And If this option is not
698 With per-numa CMA enabled, DMA users on node nid will
700 which is located in node nid, if the allocation fails,
711 which is located in node nid, if the allocation fails,
714 cmo_free_hint= [PPC] Format: { yes | no }
716 when they are freed. This is used in CMO environments
725 com20020= [HW,NET] ARCnet - COM20020 chipset
729 com90io= [HW,NET] ARCnet - COM90xx chipset (IO-mapped buffers)
733 ARCnet - COM90xx chipset (memory-mapped buffers)
742 the console buffer is full. In this case the
747 terminal emulator is active. If no 3270 terminal
748 emulator is used, this parameter has no effect.
762 See Documentation/admin-guide/serial-console.rst for more
772 Start an early, polled-mode console on the 8250/16550
775 MMIO inter-register address stride is either 8-bit
776 (mmio), 16-bit (mmio16), or 32-bit (mmio32).
778 to be equivalent to 'mmio'. 'options' are specified in
780 the h/w is not re-initialized.
799 By default we print messages on consoles in
807 syscall, or to executing "dmesg -S --raw" or to reading
810 consoleblank= [KNL] The console blank (screen saver) timeout in
831 disable the cpuidle sub-system
837 disable the cpufreq sub-system
841 policy to use. This governor must be registered in the
845 [X86] Delay for N microsec between assert and de-assert
847 on every CPU online, such as boot, and resume from suspend.
854 the parameter has no effect.
857 Run kdump after running panic-notifiers and dumping
859 succeeds in any situation.
870 [KNL, X86-64, ARM64, RISCV] Select a region under 4G first, and
873 See Documentation/admin-guide/kdump/kdump.rst for further details.
877 in the running system. The syntax of range is
878 start-[end] where start and end are both
880 Documentation/admin-guide/kdump/kdump.rst for an example.
883 [KNL, X86-64, ARM64, RISCV] range could be above 4G.
890 [KNL, X86-64, ARM64, RISCV] range under 4G. When crashkernel=X,high
895 low memory is needed to make sure DMA buffers for 32-bit
899 --> x86: max(swiotlb_size_or_default() + 8MiB, 256MiB)
900 --> arm64: 128MiB
901 --> riscv: 128MiB
909 [KNL] Disable crypto self-tests
917 csdlock_debug= [KNL] Enable or disable debug add-ons of cross-CPU
920 in case a hanging CPU is detected, and that
921 CPU is pinged again in order to try to resolve
932 See also Documentation/input/devices/joystick-parport.rst
937 [KNL] Enable printing [hashed] pointers early in the
945 [KNL] verbose locking self-tests
948 self-tests.
950 (no extra messages), setting it to -1 (all bits set)
951 will print _a_lot_ more information - normally only
965 memory corruption problems caused by bugs in kernel or
986 Format: { on, no-mount, off }
988 no-mount:
993 get a -EPERM as result when trying to register files
996 debugfs was not enabled in the kernel at all.
997 Default value is set in build-time with a kernel configuration.
1004 APIs. In addition, this is the default hugetlb size
1009 Documentation/admin-guide/mm/hugetlbpage.rst.
1013 [KNL] Debugging option to set a timeout in seconds for
1016 drivers) that have opted in will be ignored. A timeout
1023 delayacct [KNL] Enable per-task delay accounting
1036 [HW] Report power status in /proc/i8k
1053 off: No s390 zlib hardware support
1118 entries for DMA-API debugging code. One entry is
1119 required per DMA-API allocation. Use this if the
1120 DMA-API debugging code disables itself because the
1124 With this option the DMA-API debugging driver
1134 kernel data values previously stored in floating point
1144 are enabled. In order to disable RFDS mitigation all
1148 Documentation/admin-guide/hw-vuln/reg-file-data-sampling.rst
1159 panels may send no or incorrect EDID data sets.
1161 in the /lib/firmware directory that are used instead.
1162 Generic built-in EDID data sets are used, if one of
1165 and no file with the same name exists. Details and
1167 available in Documentation/admin-guide/edid.rst. An EDID
1172 data set with no connector name will be used for
1179 Control how the dt_cpu_ftrs device-tree binding is
1189 what data is available or for reverse-engineering.
1194 Documentation/admin-guide/dynamic-debug-howto.rst
1198 Enable debug messages in early_ioremap support. This
1204 When used with no options, the early console is
1205 determined by stdout-path property in device tree's
1210 Start an early, polled-mode console on a Cadence
1221 Start an early, polled-mode console on the 8250/16550
1223 MMIO inter-register address stride is either 8-bit
1224 (mmio) or 32-bit (mmio32 or mmio32be).
1227 in the same format described for "console=ttyS<n>"; if
1234 Start an early, polled-mode console on a pl011 serial
1238 the driver will use only 32-bit accessors to read/write
1247 Start an early, polled-mode console on a meson serial
1253 Start an early, polled-mode console on an msm serial
1259 Start an early, polled-mode console on an msm serial
1265 Start an early, polled-mode console on a serial port
1271 Start an early, polled-mode console on a serial port
1277 Use RISC-V SBI (Supervisor Binary Interface) for early
1295 Start an early, polled-mode console on a lantiq serial
1309 Start an early, polled-mode, output-only console on the
1314 Start an early, polled-mode console on the
1320 Start an early, polled-mode console on a Qualcomm
1328 coherent non-x86 systems that use system memory for
1364 You can find the port for a given device in
1374 The xen option can only be used in Xen domains.
1399 PCI bridges while in the EFI boot stub.
1400 nochunk: disable reading files in "chunks" in the EFI
1412 on all PCI bridges while in the EFI boot stub
1428 attribute is added to range 0x100000000-0x180000000 and
1429 0x10a0000000-0x1120000000.
1433 range 0x240000000-0x43fffffff.
1445 Documentation/admin-guide/acpi/ssdt-overlays.rst for details.
1454 This is designed to be used in conjunction with
1457 This parameter works in place of the kgdboc parameter
1459 very early in the boot process. For early debugging
1462 elanfreq= [X86-32]
1463 See comment before function elanfreq_setup() in
1470 See Documentation/admin-guide/kdump/kdump.rst for details.
1480 (in particular on some ATI chipsets).
1486 0 -- permissive (log only, no denials).
1487 1 -- enforcing (deny and log).
1519 See also Documentation/fault-injection/.
1523 See Documentation/admin-guide/sysctl/net.rst for
1527 See Documentation/admin-guide/blockdev/floppy.rst.
1530 [IA-64] Avoid check_sal_cache_flush which may hang on
1535 forcepae [X86-32]
1544 as early as possible in order to facilitate early
1566 If no parameter is passed, ftrace will dump
1571 ftrace_filter=[function-list]
1573 tracer at boot up. function-list is a comma-separated
1575 time by the set_ftrace_filter file in the debugfs
1578 ftrace_notrace=[function-list]
1579 [FTRACE] Do not trace the functions specified in
1580 function-list. This list can be changed at run time
1581 by the set_ftrace_notrace file in the debugfs
1584 ftrace_graph_filter=[function-list]
1587 function-list is a comma-separated list of functions
1589 set_graph_function file in the debugfs tracing directory.
1591 ftrace_graph_notrace=[function-list]
1592 [FTRACE] Do not trace from the functions specified in
1593 function-list. This list is a comma-separated list of
1595 set_graph_notrace file in the debugfs tracing directory.
1601 in the tracefs tracing directory. default: 0 (no limit)
1610 suspend/resume & runtime PM (consumers first, then
1613 off -- Don't create device links from firmware info.
1614 permissive -- Create device links from firmware info
1617 on -- Create device links from firmware info and use it
1618 to enforce probe and suspend/resume ordering.
1619 rpm -- Like "on", but also use to order runtime PM.
1632 strict -- Default. Continue waiting on consumers to
1634 timeout -- Give up waiting on consumers and call
1644 See also Documentation/input/devices/joystick-parport.rst
1648 gart_fix_e820= [X86-64] disable the fix e820 for K8 GART
1658 previously stored in vector registers.
1660 This issue is mitigated by default in updated microcode.
1666 microcode mitigation. No effect if the microcode
1667 mitigation is present. Known to cause crashes in
1672 gcov_persist= [GCOV] When non-zero (default), profiling data for
1682 gpio-mockup.gpio_mockup_ranges
1685 gpio-mockup.gpio_mockup_named_lines
1717 against bounds-checking flaws in the kernel's
1723 [KNL] Should the hard-lockup detector generate
1729 for 64-bit NUMA, off otherwise.
1732 hcl= [IA-64] SGI's Hardware Graph compatibility layer
1739 corresponding firmware-first mode error processing
1746 no Disable hibernation and resume.
1749 during restoration read-only).
1752 size of <nn>. This works even on boxes that have no
1770 has been explicitly set, resulting in the calling
1775 hpet= [X86-32,HPET] option to control HPET usage
1792 number of pages to allocate per-node can be specified.
1793 See also Documentation/admin-guide/mm/hugetlbpage.rst.
1798 [HW] The size of the HugeTLB pages. This is used in
1804 Documentation/admin-guide/mm/hugetlbpage.rst.
1815 boot-time allocation of gigantic hugepages is skipped.
1843 by the CONFIG_BOOTPARAM_HUNG_TASK_PANIC build-time
1849 hvc_iucv_allow= [S390] Comma-separated list of z/VM user IDs.
1866 (disabled by default, and as a pre-condition
1868 i8042.direct [HW] Put keyboard port into non-translated mode
1882 suspend-to-ram transitions, only during s2r
1906 -1 -- never invert brightness
1907 0 -- machine default
1908 1 -- force brightness inversion
1921 In such case C2/C3 won't be used again.
1922 idle=nomwait: Disable mwait for CPU C-states
1950 legacy only accept legacy-NaN binaries, if supported
1952 2008 only accept 2008-NaN binaries, if supported
1958 encodings, so if no FPU hardware is present or it has
1961 'relaxed' straps the emulator for both legacy-NaN and
1962 2008-NaN, whereas 'strict' enables legacy-NaN only on
1971 Ignore loglevel setting - this will print /all/
2003 in crypto/hash_info.h.
2038 Formats: { "ima" | "ima-ng" | "ima-ngv2" | "ima-sig" |
2039 "ima-sigv2" }
2040 Default: "ima-ng"
2072 initcall_blacklist= [KNL] Do not execute a comma-separated list of
2073 initcall functions. Useful for debugging built-in
2091 load the initrd. If an initrd is compiled in or
2092 specified in the bootparams, it takes priority over this
2109 override in debugfs after boot.
2118 0 -- basic integrity auditing messages. (Default)
2119 1 -- additional integrity auditing messages.
2129 bypassed by not enabling DMAR with this option. In
2147 could harm performance of some high-throughput
2156 1 to 9 specify maximum depth of C-state.
2165 algorithms for p-state selection. There are two
2166 P-state selection algorithms provided by
2167 intel_pstate in the active mode: powersave and
2169 on whether or not the hardware managed P-states
2170 (HWP) feature has been enabled in the processor
2176 used along with the hardware-managed P-states (HWP)
2180 in favor of acpi-cpufreq. Forcing the intel_pstate driver
2181 instead of acpi-cpufreq may disable platform features, such
2183 P-States information being indicated to OSPM and therefore
2186 or on platforms that use pcc-cpufreq instead of acpi-cpufreq.
2199 Allow per-logical-CPU P-State performance control limits using
2202 intremap= [X86-64, Intel-IOMMU]
2207 BIOS x2APIC opt-out request will be ignored
2231 0 - Try to allocate a 32-bit DMA address first, before
2233 1 - Allocate directly from the full usable range,
2235 greater than 32-bit addressing.
2239 0 - Lazy mode.
2245 1 - Strict mode.
2248 unset - Use value of CONFIG_IOMMU_DEFAULT_DMA_{LAZY,STRICT}.
2250 legacy driver-specific options takes precedence.
2255 0 - Use IOMMU translation for DMA.
2256 1 - Bypass the IOMMU for DMA.
2257 unset - Use value of CONFIG_IOMMU_DEFAULT_PASSTHROUGH.
2259 io7= [HW] IO7 for Marvel-based Alpha systems
2260 See comment before marvel_specify_io7 in
2271 No delay
2274 See Documentation/admin-guide/nfs/nfsroot.rst.
2294 to let secondary kernels in charge of setting up
2298 Enables support for pseudo-NMIs in the kernel. This
2317 [Deprecated - use cpusets instead]
2318 Format: [flag-list,]<cpu-list>
2321 specified in the flag list (default: domain):
2324 Disable the tick when a single task runs.
2326 A residual 1Hz tick is offloaded to workqueues, which you
2344 move in and out of an isolated set anytime.
2349 "number of CPUs in system - 1".
2368 CPUs then this parameter has no effect on the
2372 housekeeping CPUs has no influence on those
2375 The format of <cpu-list> is described above.
2379 ivrs_ioapic [HW,X86-64]
2380 Provide an override to the IOAPIC-ID<->DEVICE-ID
2381 mapping provided in the IVRS ACPI table.
2384 For example, to map IOAPIC-ID decimal 10 to
2390 * To map IOAPIC-ID decimal 10 to PCI device 00:14.0
2393 * To map IOAPIC-ID decimal 10 to PCI segment 0x1 and
2397 ivrs_hpet [HW,X86-64]
2398 Provide an override to the HPET-ID<->DEVICE-ID
2399 mapping provided in the IVRS ACPI table.
2402 For example, to map HPET-ID decimal 10 to
2408 * To map HPET-ID decimal 0 to PCI device 00:14.0
2411 * To map HPET-ID decimal 10 to PCI segment 0x1 and
2415 ivrs_acpihid [HW,X86-64]
2416 Provide an override to the ACPI-HID:UID<->DEVICE-ID
2417 mapping provided in the IVRS ACPI table.
2420 For example, to map UART-HID:UID AMD0020:0 to
2426 * To map UART-HID:UID AMD0020:0 to PCI segment is 0,
2429 * To map UART-HID:UID AMD0020:0 to PCI segment 0x1 and
2444 useful for debugging when something happens in the window
2450 kernelcore= [KNL,X86,IA-64,PPC]
2453 the kernel for non-movable allocations. The requested
2454 amount is spread evenly throughout all nodes in the
2456 movable memory in its own zone, ZONE_MOVABLE. In the
2463 subsystem. Note that allocations like PTEs-from-HighMem
2467 It is possible to specify the exact amount of memory in
2469 memory in the form of "nn%", or "mirror". If "mirror"
2471 for non-movable allocations and remaining memory is used
2479 optional and is the number seconds in between
2480 each poll cycle to the debug port in case you need
2482 gdb or control-c on the dbgp connection. When
2483 not using this parameter you use sysrq-g to break into
2488 or a supported polling keyboard driver (non-usb).
2498 characters and can work in polling mode, you can use
2515 Configure the RouterBoard 532 series on-chip
2518 kmemleak= [KNL] Boot-time kmemleak enable/disable
2524 kprobe_event=[probe-list]
2526 The probe-list is a semicolon delimited list of probe
2556 execution by eliminating the write-protection faults
2567 disabled, all huge pages in a memslot will be eagerly
2601 back to huge pages. If the value is a non-zero N, KVM will
2606 kvm-amd.nested= [KVM,AMD] Control nested virtualization feature in
2609 kvm-amd.npt= [KVM,AMD] Control KVM's use of Nested Page Tables,
2610 a.k.a. Two-Dimensional Page Tables. Default is 1
2614 kvm-arm.mode=
2619 nvhe: Standard nVHE-based mode, without support for
2622 protected: nVHE-based mode with support for guests whose
2625 nested: VHE-based mode with support for nested
2634 kvm-arm.vgic_v3_group0_trap=
2635 [KVM,ARM] Trap guest accesses to GICv3 group-0
2638 kvm-arm.vgic_v3_group1_trap=
2639 [KVM,ARM] Trap guest accesses to GICv3 group-1
2642 kvm-arm.vgic_v3_common_trap=
2646 kvm-arm.vgic_v4_enable=
2658 kvm-intel.ept= [KVM,Intel] Control KVM's use of Extended Page Tables,
2659 a.k.a. Two-Dimensional Page Tables. Default is 1
2663 kvm-intel.emulate_invalid_guest_state=
2665 state. Ignored if kvm-intel.enable_unrestricted_guest=1,
2671 kvm-intel.flexpriority=
2676 kvm-intel.nested=
2677 [KVM,Intel] Control nested virtualization feature in
2680 kvm-intel.unrestricted_guest=
2686 kvm-intel.vmentry_l1d_flush=[KVM,Intel] Mitigation for L1 Terminal Fault
2687 CVE-2018-3620.
2696 Default is cond (do L1 cache flush in specific instances)
2698 kvm-intel.vpid= [KVM,Intel] Control KVM's use of Virtual Processor
2710 In vulnerable processors, the speculatively
2711 forwarded data can be used in a cache side channel
2718 on - enable the interface for the mitigation
2729 enables all mitigations in the
2735 when the first VM is started in a
2753 when the first VM is started in a
2765 when the first VM is started in a
2771 warn when a VM is started in a potentially
2783 For details see: Documentation/admin-guide/hw-vuln/l1tf.rst
2789 lapic [X86-32,APIC] Enable the local APIC even if BIOS
2793 value for LAPIC timer one-shot implementation. Default
2794 back to the programmable timer unit in the LAPIC.
2798 in C2 power state.
2812 libata.noacpi [LIBATA] Disables use of ACPI in libata suspend/resume
2816 libata.force= [LIBATA] Force configurations. The format is a comma-
2828 first fan-out link behind PMP device. It does not
2833 as there is no ambiguity, shortcut notation is allowed.
2842 * Transfer mode: pio[0-7], mwdma[0-4] and udma[0-7].
2849 * rstonce: only attempt one reset during hot-unplug
2852 * [no]dbdelay: Enable or disable the extra 200ms delay
2856 * [no]ncq: Turn on or off NCQ.
2858 * [no]ncqtrim: Enable or disable queued DSM TRIM.
2860 * [no]ncqati: Enable or disable NCQ trim on ATI chipset.
2862 * [no]trim: Enable or disable (unqueued) TRIM.
2868 * [no]dma: Turn on or off DMA transfers.
2875 * [no]dmalog: Enable or disable the use of the
2878 * [no]iddevlog: Enable or disable access to the
2881 * [no]logdir: Enable or disable access to the general
2892 * [no]lpm: Enable or disable link power management.
2894 * [no]setxfer: Indicate if transfer speed mode setting
2897 * [no]fua: Disable or enable FUA (Force Unit Access)
2931 Set the number of locking read-acquisition kthreads.
2936 Set the number of locking write-acquisition kthreads.
2939 Set time (s) after boot for CPU-hotplug testing.
2942 Set time (s) between CPU-hotplug operations, or
2943 zero to disable CPU-hotplug testing.
2946 Set task-shuffle interval (jiffies). Shuffling
2947 tasks allows some CPUs to go into dyntick-idle
2952 is useful for hands-off automated testing.
2968 Run the write-side locktorture kthreads at
2969 sched_set_fifo() real-time priority.
2989 7 (KERN_DEBUG) debug-level messages
2992 in bytes. n must be a power of two and greater
2999 logo.nologo [FB] Disables display of the built-in Linux logo.
3008 specified in addition to the ports) causes
3017 an IEEE 1284-compliant printer is attached; if
3023 time-consuming boot-time autodetection (up to 250 ms per
3030 will cause delays in the kernel to be wrong, leading to
3032 unlikely, in the extreme case this might damage your
3044 machvec= [IA-64] Force the use of a particular machine-vector
3045 (machvec) in a generic kernel.
3050 Example: machtype=lemote-yeeloong-2f-7inch
3052 max_addr=nn[KMG] [KNL,BOOT,IA-64] All physical memory greater
3065 (loop.max_loop) unconditionally pre-created at init time. The default
3068 devices can be requested on-demand with the
3069 /dev/loop-control interface.
3071 mce [X86-32] Machine Check Exception
3073 mce=option [X86-64] See Documentation/arch/x86/x86_64/boot-options.rst
3076 See Documentation/admin-guide/md.rst.
3083 Control mitigation for the Micro-architectural Data
3090 In vulnerable processors, the speculatively
3091 forwarded data can be used in a cache side channel
3098 full - Enable MDS mitigation on vulnerable CPUs
3099 full,nosmt - Enable MDS mitigation and disable
3101 off - Unconditionally disable MDS mitigation
3103 On TAA-affected machines, mds=off can be prevented by
3105 mitigated with the same mechanism so in order to disable
3112 For details see: Documentation/admin-guide/hw-vuln/mds.rst
3118 Amount of memory to be used in cases as follows:
3126 [ARC,MICROBLAZE] - the limit applies only to low memory,
3129 [ARM64] - only limits memory covered by the linear
3138 in above case 3, memory may need be hot added after boot
3142 [ARM,MIPS] - override the memory layout reported by
3149 mem=nopentium [BUGS=X86-32] Disable usage of 4MB pages for kernel
3156 per-device physically contiguous DMA buffers.
3164 See Documentation/admin-guide/mm/memory-hotplug.rst.
3189 Example: Exclude memory from 0x18690000-0x1869ffff
3203 memmap=<size>%<offset>-<oldtype>+<newtype>
3205 from <oldtype> to <newtype>. If "-<oldtype>" is left
3214 memory when doing things like suspend/resume.
3220 repeatable BIOS-originated corruption always
3225 By default it checks for corruption in the low
3228 corruption in more or less memory.
3247 allocations in some configurations (e.g. small
3249 The state of the flag can be read in
3264 mem_encrypt= [X86-64] AMD Secure Memory Encryption (SME) control
3270 Refer to Documentation/virt/kvm/x86/amd-memory-encryption.rst
3273 mem_sleep_default= [SUSPEND] Default system suspend mode:
3274 s2idle - Suspend-To-Idle
3275 shallow - Power-On Suspend or equivalent (if supported)
3276 deep - Suspend-To-RAM or equivalent (if supported)
3277 See Documentation/admin-guide/pm/sleep-states.rst.
3282 mfgpt_irq= [IA-32] Specify the IRQ to use for the
3283 Multi-Function General Purpose Timers on AMD Geode
3286 mfgptfix [X86-32] Fix MFGPT timers on AMD Geode platforms when
3298 min_addr=nn[KMG] [KNL,BOOT,IA-64] All physical memory below this
3305 0 - The attached screen is the 3.5" TFT
3306 1 - The attached screen is the 7" TFT
3307 2 - The VGA Shield is attached (1024x768)
3311 b - Enable backlight. The TFT backlight pin will be
3315 c - Enable the s3c camera interface.
3316 t - Reserved for enabling touchscreen support. The
3317 touchscreen support is not enabled in the mainstream
3319 in the "bleeding edge" mini2440 support kernel at
3320 https://repo.or.cz/w/linux-2.6/mini2440.git
3325 arch-independent options, each of which is an
3326 aggregation of existing arch-specific options.
3354 ssbd=force-off [ARM64]
3367 have other ways of avoiding SMT-based attacks.
3394 operation. Exposed data could originate or end in
3402 full - Enable mitigation on vulnerable CPUs
3404 full,nosmt - Enable mitigation and disable SMT on
3407 off - Unconditionally disable mitigation
3412 mitigated with the same mechanism so in order to
3420 Documentation/admin-guide/hw-vuln/processor_mmio_stale_data.rst
3423 If no <bool> value is specified or if the value
3450 module_blacklist= [KNL] Do not load a comma-separated list of
3457 touchpads working in absolute mode only).
3464 movablecore= [KNL,X86,IA-64,PPC]
3475 movable_node [KNL] Boot-time switch to make hotplugable memory
3482 Format: <name>,<region-number>,<size>,<offset>
3485 <name>,<region-number>[,<base>,<size>,<buswidth>,<altbuswidth>]
3493 See arch/arm/mach-s3c/mach-jive.c
3529 something different and driver-specific.
3530 This usage is only documented in each driver source
3534 [NET] Specifies amount of time (in seconds) that
3562 [NFS] enable 64-bit inode numbers.
3563 If zero, the NFS client will fake up a 32-bit inode
3565 of returning the full 64-bit number.
3566 The default is to return 64-bit inode numbers.
3576 callbacks the client will process in parallel for
3584 Note that there is little point in setting this
3592 'sec=sys' security flavour. In effect it is
3601 [NFS4] Specify an additional fixed unique ident-
3610 no guarantees that the file will remain unchanged
3620 information in exchange_id requests.
3621 If zero, no implementation identification information
3632 driver. A non-zero value sets the minimum interval
3633 in seconds between layoutstats transmissions.
3637 server-to-server copies for which this server is
3649 server-to-server copy, knfsd temporarily mounts
3650 the source server. It caches the mount in case
3656 See Documentation/admin-guide/nfs/nfsroot.rst.
3658 nfsroot= [NFS] nfs root filesystem for disk-less boxes.
3659 See Documentation/admin-guide/nfs/nfsroot.rst.
3662 See Documentation/admin-guide/nfs/nfsroot.rst.
3665 Dump stacks even of idle CPUs in response to an
3666 NMI stack-backtrace request.
3675 0 - turn hardlockup detector in nmi_watchdog off
3676 1 - turn hardlockup detector in nmi_watchdog on
3688 no387 [BUGS=X86-32] Tells the kernel to use the 387 maths
3692 no4lvl [RISCV] Disable 4-level and 5-level paging modes. Forces
3693 kernel to use 3-level paging instead.
3695 no5lvl [X86-64,RISCV] Disable 5-level paging mode. Forces
3696 kernel to use 4-level paging instead.
3699 caches in the slab allocator. Saves per-node memory,
3708 IOAPICs that may be present in the system.
3715 [HW] Never suspend the console
3716 Disable suspending of consoles during suspend and
3720 debugging driver suspend/resume hooks). This may
3736 no_entry_flush [PPC] Don't flush the L1-D cache when entering the kernel.
3738 noexec [IA-64]
3740 noexec32 [X86-64]
3741 This affects only 32-bit executables.
3742 noexec32=on: enable non-executable mappings (default)
3744 noexec32=off: disable non-executable mappings
3755 nofxsr [BUGS=X86-32] Disables x86 floating point extended
3757 legacy floating-point registers on task switch.
3759 nohalt [IA-64] Tells the kernel not to use the power saving
3761 power-consumption. On the positive side, it reduces
3762 interrupt wake-up latency, which may improve performance
3763 in certain environments such as networked servers or
3764 real-time systems.
3773 difficult since unequal pointers can no longer be
3774 compared. However, if this command-line option is
3783 busy wait in do_idle() and not use the arch_cpu_idle()
3801 In kernels built with CONFIG_NO_HZ_FULL=y, set
3802 the specified list of CPUs whose tick will be stopped
3805 in this list will have their RCU callbacks offloaded,
3806 just as if they had also been called out in the
3815 nointremap [X86-64, Intel-IOMMU] Do not enable interrupt
3817 [Deprecated - use intremap=off]
3819 nointroute [IA-64]
3825 noirqdebug [X86-32] Disables the code which attempts to detect and
3830 nojitter [IA-64] Disables jitter checking for ITC timers.
3837 no-kvmapf [X86,KVM] Disable paravirtualized asynchronous page
3840 no-kvmclock [X86,KVM] Disable paravirtualized KVM clock driver
3842 nolapic [X86-32,APIC] Do not enable or use the local APIC.
3844 nolapic_timer [X86-32,APIC] Do not use the local APIC timer.
3846 nomca [IA-64] Disable machine check abort handling
3848 nomce [X86-32] Disable Machine Check Exception
3850 nomfgpt [X86-32] Disable Multi-Function General Purpose
3856 not load if they could possibly displace the pre-
3859 perform display-mode changes or accelerated rendering.
3872 nopcid [X86-64] Disable the PCID cpu feature.
3875 in some Intel CPUs.
3877 nopti [X86-64]
3882 as generic guest with no PV drivers. Currently support
3893 noreplace-smp [X86-32,SMP] Don't replace SMP instructions
3899 nosbagart [IA-64]
3901 no-scroll [VGA] Disables scrollback.
3902 This is required for the Braillex ib80-piezo Braille
3905 nosgx [X86-64,SGX] Disables Intel SGX kernel support.
3925 nosoftlockup [KNL] Disable the soft-lockup detector.
3930 nospectre_bhb [ARM64] Disable all mitigations for Spectre-BHB (branch
3936 possible in the system.
3943 no-steal-acc [X86,PV_OPS,ARM64,PPC/PSERIES] Disable paravirtualized
3953 [PPC] Don't flush the L1-D cache after accessing user data.
3959 without any limit and this data is stored in memory,
3962 data will be no longer available. This parameter
3966 no-vmw-sched-clock
3971 soft-lockup and NMI watchdog (hard-lockup).
3975 nox2apic [X86-64,APIC] Do not enable x2APIC mode.
3978 LEGACY_XAPIC_DISABLED bit set in the
3983 enabling legacy floating-point and sse state.
3985 noxsaveopt [X86] Disables xsaveopt used in saving x86 extended
3992 noxsaves [X86] Disables xsaves and xrstors used in saving and
3993 restoring x86 extended register state in compacted
3996 in standard form of xsave area. By using this
4001 This parameter sets the maximum duration, in
4009 nptcg= [IA-64] Override max number of concurrent global TLB
4016 number of already plugged CPU during bootup, later in
4018 n. So during boot up some boot time memory for per-cpu
4019 variables need be pre-allocated for later physical cpu
4034 See Documentation/admin-guide/sysctl/vm.rst for details.
4037 See Documentation/core-api/debugging-via-ohci1394.rst for more
4052 onenand.bdry= [HW,MTD] Flex-OneNAND Boundary Configuration
4056 boundary - index of last SLC block on Flex-OneNAND.
4058 lock - Configure if Flex-OneNAND boundary should be locked.
4072 running on a platform with a direct-mapped memory-side
4078 page_owner= [KNL] Boot-time page_owner enabling option.
4080 each page is disabled in default. With this switch,
4084 page_poison= [KNL] Boot-time parameter changing the state of
4102 panic_on_taint= Bitmask for conditionally calling panic() in add_taint()
4106 called with any of the flags in this set.
4111 See Documentation/admin-guide/tainted-kernels.rst for
4125 bit 5: print all printk messages in buffer
4126 bit 6: print all CPUs backtrace (if available in the arch)
4128 so there are risks of losing older messages in the log.
4149 Parallel ports are assigned in the order they
4154 Configure VIA parallel port to operate in
4156 computer where firmware has no options for setting
4163 Set to non-zero to probe primary and secondary ISA
4164 port ranges on PCI systems where no PCI PATA device
4169 Set to non-zero if a chip is present that snoops speed
4217 Set to non-zero to probe tertiary and further ISA
4227 corresponding bits in the mask to 1. Bit 0 is for
4228 the first port in the list above (0x1f0), and so on.
4233 Set to non-zero to probe QDI controllers. By default
4238 Set to non-zero to probe Winbond controllers. Use
4262 specified in one of the following formats:
4280 devices in the system.
4285 bios [X86-32] force use of PCI BIOS, don't access
4287 has a non-standard PCI host bridge.
4288 nobios [X86-32] disallow use of PCI BIOS, only direct
4293 Mechanism 1 (config address in IO port 0xCF8,
4294 data in IO port 0xCFC, both 32-bit).
4296 Mechanism 2 (IO port 0xCF8 is an 8-bit port for
4297 the function, IO port 0xCFA, also 8-bit, sets
4299 through ports 0xC000-0xCFFF).
4306 root domains (aka PCI segments, in ACPI-speak).
4314 disable the use of MSI interrupts system-wide.
4319 primary IO-APIC for bridges that cannot disable
4326 biosirq [X86-32] Use PCI BIOS calls to get the interrupt
4350 F0000h-100000h range.
4355 assign-busses [X86] Always assign all PCI bus
4359 in the BIOS $PIR table. This is needed on
4362 notebooks. This will have no effect if ACPI
4374 for BIOS defects in host bridge _CRS methods.
4376 <linux-pci@vger.kernel.org>.
4380 a bug to <linux-pci@vger.kernel.org>.
4382 This is normally done in pci_enable_device(),
4392 bfsort Sort PCI devices into breadth-first order.
4395 nobfsort Don't sort PCI devices into breadth-first order.
4397 tuning and use the BIOS-configured MPS defaults.
4403 value (no larger than the MPS that the device
4407 configuration allows peer-to-peer DMA between
4410 that hot-added devices will work.
4425 A PCI-PCI bridge can be specified if resource
4431 for 4096-byte alignment.
4433 end-to-end CRC checking). Only effective if
4464 noats [PCIE, Intel-IOMMU, AMD-IOMMU]
4476 Specify one or more PCI devices (in the format
4483 may put more devices in an IOMMU group.
4500 dpc-native Use native PCIe service for DPC only. May
4517 Keep all power-domains already enabled by bootloader on,
4518 even if no driver has claimed them. This is useful
4530 See comments in mm/percpu.c for details on each
4534 pirq= [SMP,APIC] Manual mp-table setup
4535 See Documentation/arch/x86/i386/IO-APIC.rst.
4539 See also Documentation/admin-guide/parport.rst.
4546 This option takes over the PMU facility, so it is no
4553 pm_debug_messages [SUSPEND,KNL]
4554 Enable suspend/resume debug messages during boot up.
4558 CONFIG_PNP_DEBUG_MESSAGES option). Change at run-time
4567 { on | off | curr | res | no-curr | no-res }
4576 Ranges are in pairs (I/O port base and size).
4581 Ranges are in pairs (memory base and size).
4607 none - Limited to cond_resched() calls
4608 voluntary - Limited to cond_resched() and might_sleep() calls
4609 full - Any section that isn't explicitly preempt disabled
4612 print-fatal-signals=
4618 coredump - etc.
4621 you might want to try "ulimit -i unlimited".
4633 or lockdep-detected issues (only if lock debug is on).
4636 in order to provide more debug information.
4642 on - unlimited logging to /dev/kmsg from userspace
4643 off - logging to /dev/kmsg disabled
4644 ratelimit - ratelimit the logging
4655 - 'always': traditional behavior always allows mem overrides.
4656 - 'ptrace': only allow mem overrides for active ptracers.
4657 - 'never': never allow mem overrides.
4661 Limit processor to maximum C-state
4665 Ignore the _CST method to determine C-states,
4672 Param: "schedule" - profile schedule points.
4673 Param: "kvm" - profile VM exits.
4674 Param: <number> - step/bucket size as a power of 2 for
4690 psmouse.rate= [HW,MOUSE] Set desired mouse report rate, in reports
4696 [HW,MOUSE] Set desired mouse resolution, in dpi.
4703 pti= [X86-64] Control Page Table Isolation of user and
4708 on - unconditionally enable
4709 off - unconditionally disable
4710 auto - kernel detects whether your CPU model is
4716 [KNL] Number of legacy pty's. Overwrites compiled-in
4728 See Documentation/admin-guide/md.rst.
4730 ramdisk_size= [RAM] Sizes of RAM disks in kilobytes
4731 See Documentation/admin-guide/blockdev/ramdisk.rst.
4750 cross-syscall address exposures. This is only
4756 ras=option[,option,...] [KNL] RAS-specific options
4762 rcu_nocbs[=cpu-list]
4766 In kernels built with CONFIG_RCU_NOCB_CPU=y,
4767 enable the no-callback CPU mode, which prevents
4768 such CPUs' callbacks from being invoked in
4772 "p" for RCU-preempt, "s" for RCU-sched, and "g"
4776 and real-time workloads. It can also improve
4780 list of CPUs is set to no-callback mode from boot.
4783 arguments are omitted, no CPU will be set to
4784 no-callback mode from boot but the mode may be
4795 This improves the real-time response for the
4803 process in one batch.
4812 RCU grace-period cleanup.
4816 RCU grace-period initialization.
4820 RCU grace-period pre-initialization, that is,
4821 the propagation of recent CPU-hotplug changes up
4825 Set delay from grace-period initialization to
4836 Set required age in jiffies for a
4838 soliciting quiescent-state help from
4844 This calculated value may be viewed in
4850 Set the SCHED_FIFO priority of the RCU per-CPU
4853 and for the RCU grace-period kthreads (rcu_bh,
4855 set, valid values are 1-99 and the default is 1
4856 (the least-favored priority). Otherwise, when
4857 RCU_BOOST is not set, valid values are 0-99 and
4858 the default is zero (non-realtime operation).
4863 On callback-offloaded (rcu_nocbs) CPUs,
4866 use of the ->nocb_bypass list. However, in the
4867 common non-flooded case, RCU queues directly to
4868 the main ->cblist in order to avoid the extra
4869 overhead of the ->nocb_bypass list and its lock.
4871 a single jiffy, RCU pre-queues the callbacks into
4872 the ->nocb_bypass queue. The definition of "too
4881 batch limiting is re-enabled.
4885 RCU's force-quiescent-state scan will aggressively
4893 Set the page-cache refill delay (in milliseconds)
4894 in response to low-memory conditions. The range
4895 of permitted values is in the range 0:100000.
4898 Set the shift-right count to use to compute
4899 the callback-invocation batch limit bl from
4903 callbacks, the softirq handler will exit in
4906 Please note that this callback-invocation batch
4907 limit applies only to non-offloaded callback
4909 invoked in the context of an rcuoc kthread, which
4916 cache-to-cache transfer latencies.
4922 and for NUMA systems with large remote-access
4931 whole algorithm to behave better in low memory
4935 Set the number of NOCB callback kthreads in
4938 the wakeup overhead on the global grace-period
4940 each group's NOCB grace-period kthread.
4943 Cause the grace-period kthread to get an extra
4945 it should at force-quiescent-state time.
4953 every 32 callbacks in order to limit the pain
4957 In CONFIG_RCU_STRICT_GRACE_PERIOD=y kernels,
4958 this specifies an rcu_read_unlock()-time delay
4959 in microseconds. This defaults to zero.
4962 of RCU-protected pointers after the relevant
4972 per-CPU rcuc kthreads. Defaults to a non-zero
4982 grace-period primitives such as call_rcu().
4993 grace-period primitives.
4996 Set test-start holdoff period. The purpose of
4998 test until boot completes in order to avoid
5002 In kernels built with CONFIG_RCU_LAZY=y, test
5014 Test the double-argument variant of kfree_rcu().
5016 rcuscale.kfree_rcu_test_single, both the single-
5017 and double-argument variants are tested.
5020 Test the single-argument variant of kfree_rcu().
5022 rcuscale.kfree_rcu_test_double, both the single-
5023 and double-argument variants are tested.
5029 Number of allocations and frees done in an iteration.
5036 Set the minimum test run time in seconds. This
5037 does not affect the data-collection interval,
5042 Set number of RCU readers. The value -1 selects
5044 "n" less than -1 selects N-n+1, where N is again
5045 the number of CPUs. For example, -2 selects N
5046 (the number of CPUs), -3 selects N+1, and so on.
5047 A value of "n" less than or equal to -N selects
5060 complete. This is useful for hands-off automated
5067 Write-side holdoff between grace periods,
5068 in microseconds. The default of zero says
5069 no holdoff.
5072 Additional write-side holdoff between grace
5073 periods, but in jiffies. The default of zero
5074 says no holdoff.
5078 in microseconds.
5082 in microseconds.
5086 in seconds.
5090 for RCU grace-period forward-progress testing
5097 Specify the fraction of a CPU-stall-warning
5098 period to do tight-loop forward-progress testing.
5102 forward-progress tests.
5106 need_resched() during tight-loop forward-progress
5110 Use conditional/asynchronous update-side
5114 Use expedited update-side primitives, if available.
5117 Use normal (non-expedited) asynchronous
5118 update-side primitives, if available.
5121 Use normal (non-expedited) synchronous
5122 update-side primitives, if available. If all
5126 they are all non-zero.
5134 Leak an RCU-protected pointer out of the reader.
5135 This can of course result in splats, and is
5145 stress RCU, they don't participate in the actual
5149 Set number of RCU callback-offload togglers.
5153 Set the delay in milliseconds between successive
5154 callback-offload toggling attempts.
5157 Set number of RCU readers. The value -1 selects
5158 N-1, where N is the number of CPUs. A value
5159 "n" less than -1 selects N-n-2, where N is again
5160 the number of CPUs. For example, -2 selects N
5161 (the number of CPUs), -3 selects N+1, and so on.
5164 Enable debug-object double-call_rcu() testing.
5167 Set time (s) after boot for CPU-hotplug testing.
5170 Set time (jiffies) between CPU-hotplug operations,
5171 or zero to disable CPU-hotplug testing.
5174 Set the number of read-then-exit kthreads used
5176 task-exit processing.
5179 The number of times in a given read-then-exit
5180 episode that a set of read-then-exit kthreads
5184 The delay, in seconds, between successive
5185 read-then-exit testing episodes.
5188 Set task-shuffle interval (s). Shuffling tasks
5189 allows some CPUs to go into dyntick-idle mode
5194 is useful for hands-off automated testing.
5202 in warnings from preemptible RCU in addition to
5203 any other stall-related activity. Note that
5204 in kernels built with CONFIG_PREEMPTION=n and
5209 in scheduling-while-atomic splats.
5211 Use of this module parameter results in splats.
5222 grace-period kthread to test RCU CPU stall
5237 Test RCU priority boosting? 0=no, 1=maybe, 2=yes.
5248 Test RCU's dyntick-idle handling. See also the
5272 The value is in seconds and the maximum allowed
5277 messages. The value is in milliseconds
5280 adjusted to an arch timer tick resolution.
5297 Use expedited grace-period primitives, for
5301 real-time latency, and degrade energy efficiency.
5302 No effect on CONFIG_TINY_RCU kernels.
5305 Use only normal grace-period primitives,
5308 real-time latency, CPU utilization, and
5310 increased grace-period latency. This parameter
5311 overrides rcupdate.rcu_expedited. No effect on
5317 only normal grace-period primitives. No effect
5323 post-boot attempt at an expedited RCU grace
5324 period to instead use normal non-expedited
5325 grace-period processing.
5333 set to the default value of -1.
5336 Set the minimum number of callback-queuing-time
5337 lock-contention events per jiffy required to
5338 cause the RCU Tasks flavors to switch to per-CPU
5341 the default value of -1.
5346 of -1 allows this to be automatically (and
5348 for use in testing.
5351 Set time in jiffies during which RCU tasks will
5354 number avoids disturbing real-time workloads,
5359 cancel laziness on that CPU. Use -1 to disable
5365 Set initial timeout in jiffies for RCU task stall
5369 only printed prior to the stall-warning message
5372 seconds. A change in value does not take effect
5387 Set timeout in jiffies for RCU task stall
5390 A change in value does not take effect until
5394 Set timeout in milliseconds RCU Tasks asynchronous
5401 Set timeout in milliseconds RCU Tasks
5409 Set timeout in milliseconds RCU Tasks
5425 force - Override the decision by the kernel to hide the
5428 support, specifically around the suspend/resume
5453 Set test-start holdoff period. The purpose of
5455 test until boot completes in order to avoid
5462 but the default has already reduced the per-pass
5467 Set number of readers. The default value of -1
5476 Set the read-side critical-section duration,
5477 measured in microseconds.
5480 Specify the read-protection implementation to test.
5499 See Documentation/admin-guide/cgroup-v1/cpusets.rst.
5507 reservetop= [X86-32]
5516 Specify the partition device for software suspend
5523 in <PAGE_SIZE> units (needed only for swap files).
5524 See Documentation/power/swsusp-and-swap-files.rst
5526 resumedelay= [HIBERNATION] Delay (in seconds) to pause before attempting to
5539 AMD-based UNRET and IBPB mitigations alone do not stop
5541 sibling threads. For that reason, STIBP is used on pro-
5545 off - no mitigation
5546 auto - automatically select a migitation
5547 auto,nosmt - automatically select a mitigation,
5551 ibpb - On AMD, mitigate short speculation
5556 ibpb,nosmt - Like "ibpb" above but will disable SMT
5560 unret - Force enable untrained return thunks,
5561 only effective on AMD f15h-f17h based
5563 unret,nosmt - Like unret, but will disable SMT when STIBP
5598 ro [KNL] Mount root device read-only on boot
5601 on Mark read-only kernel memory as read-only (default).
5602 off Leave read-only kernel memory writable for debugging.
5603 full Mark read-only kernel memory and aliases as read-only
5609 debug-uart get routed to the D+ and D- pins of the usb
5614 see the early_lookup_bdev comment in
5615 block/early-lookup.c for details.
5620 rootdelay= [KNL] Delay (in seconds) to pause before attempting to
5631 rootwait= [KNL] Maximum time (in seconds) to wait for root device
5640 rw [KNL] Mount root device read-write on boot
5642 S [KNL] Run init in single mode
5647 With strict flushing every unmap operation will result in
5659 and so on. As a special case a factor of 0 imposes no
5670 incurs a small amount of overhead in the scheduler
5692 to 10 seconds for built-in smp_call_function()
5697 up to the chosen limit in seconds. Zero (the
5699 that requesting even small non-zero numbers of
5700 seconds can result in RCU CPU stall warnings,
5706 The default of -1 specifies a number of kthreads
5711 test before initiating CPU-hotplug operations.
5715 CPU-hotplug operations. Specifying zero (which
5716 is the default) disables CPU-hotplug operations.
5722 Non-zero values are useful for automated tests.
5745 "wait" parameter. A value of -1 selects the
5746 default if all other weights are -1. However,
5748 value of -1 will instead select a weight of zero.
5753 non-zero "wait" parameter. See weight_single.
5766 non-zero "wait" parameter. See weight_single
5778 non-zero "wait" parameter. See weight_single
5781 skew_tick= [KNL] Offset the periodic timer tick per cpu to mitigate
5785 0 -- disable. (may be 1 via CONFIG_CMDLINE="skew_tick=1"
5786 1 -- enable.
5797 0 -- disable.
5798 1 -- enable.
5801 serialnumber [BUGS=X86-32]
5803 sev=option[,option...] [X86-64] See Documentation/arch/x86/x86_64/boot-options.rst
5811 to set it to "all" by meaning -- no limit here.
5817 simeth= [IA-64]
5829 allocs to different slabs, especially in hardened
5849 may poison objects when not in use. Also tracks the
5884 Specify the period of time in milliseconds
5892 using the csdlock_debug- kernel parameter.
5897 system. By default, let CSD-lock acquisition
5899 for this value provides a 5-minute timeout.
5901 smsc-ircc2.nopnp [HW] Don't use PNP to discover SMC devices
5902 smsc-ircc2.ircc_cfg= [HW] Device configuration I/O port
5903 smsc-ircc2.ircc_sir= [HW] SIR base I/O port
5904 smsc-ircc2.ircc_fir= [HW] FIR base I/O port
5905 smsc-ircc2.ircc_irq= [HW] IRQ line
5906 smsc-ircc2.ircc_dma= [HW] DMA channel
5907 smsc-ircc2.ircc_transceiver= [HW] Transceiver type:
5917 Default: -1 (no limit)
5920 [KNL] Should the soft-lockup detector generate panics.
5923 A value of 1 instructs the soft-lockup detector
5924 to panic the machine when a soft-lockup occurs. It is
5927 respective build-time switch to that functionality.
5930 [KNL] Should the soft-lockup detector generate
5935 See Documentation/admin-guide/laptops/sonypi.rst
5942 on - (default) Enable the HW or SW mitigation
5944 off - Disable the mitigation.
5951 on - unconditionally enable, implies
5953 off - unconditionally disable, implies
5955 auto - kernel detects whether your CPU model is
5972 retpoline - replace indirect branches
5973 retpoline,generic - Retpolines
5974 retpoline,lfence - LFENCE; indirect branch
5975 retpoline,amd - alias for retpoline,lfence
5976 eibrs - Enhanced/Auto IBRS
5977 eibrs,retpoline - Enhanced/Auto IBRS + Retpolines
5978 eibrs,lfence - Enhanced/Auto IBRS + LFENCE
5979 ibrs - use IBRS to protect kernel
5989 on - Unconditionally enable mitigations. Is
5992 off - Unconditionally disable mitigations. Is
5995 prctl - Indirect branch speculation is enabled,
6001 - Like "prctl" above, but only STIBP is
6007 - Same as "prctl" above, but all seccomp
6012 - Like "seccomp" above, but only STIBP is
6017 auto - Kernel selects the mitigation depending on
6028 off - Disable mitigation
6029 microcode - Enable microcode mitigation only
6030 safe-ret - Enable sw-only safe RET mitigation (default)
6031 ibpb - Enable mitigation by issuing IBPB on
6033 ibpb-vmexit - Issue IBPB only on VMEXIT
6034 (cloud-specific mitigation)
6042 as "Speculative Store Bypass" in which recent stores
6049 In vulnerable processors, the speculatively forwarded
6050 store can be used in a cache side channel attack, for
6059 on - Unconditionally disable Speculative Store Bypass
6060 off - Unconditionally enable Speculative Store Bypass
6061 auto - Kernel detects whether the CPU model contains an
6067 prctl - Control Speculative Store Bypass per thread
6071 seccomp - Same as "prctl" above, but all seccomp threads
6079 on,auto - On Power8 and Power9 insert a store-forwarding
6083 off - No action.
6098 boundaries will result in an alignment check exception
6102 off - not enabled
6104 warn - the kernel will emit rate-limited warnings
6110 enabled in hardware.
6112 fatal - the kernel will send SIGBUS to applications
6115 both features are enabled in hardware.
6117 ratelimit:N -
6125 If an #AC exception is hit in the kernel or in
6126 firmware (i.e. not while executing in user mode)
6127 the kernel will oops in either "warn" or "fatal"
6137 Certain CPUs are vulnerable to an MDS-like
6145 result in reduced throughput from /dev/urandom.
6157 This kernel-boot parameter defaults to 128,
6158 but takes effect only when the low-order four
6175 instead of the compile-time CONFIG_NR_CPUS.
6179 grace-period sequence counter wrap for the
6180 srcu_data structure's ->srcu_gp_seq_needed field.
6181 The greater the number of bits set in this kernel
6195 Specifies the number of no-delay instances
6202 Specifies the per-grace-period phase, number of
6203 non-sleeping polls of readers. Beyond this limit,
6209 Specifies number of microseconds of non-sleeping
6210 delay between each non-sleeping poll of readers.
6213 Specifies the number of update-side contention
6218 set for contention-based conversions to occur.
6228 force-on: Unconditionally enable mitigation for
6230 force-off: Unconditionally disable mitigation for
6232 kernel: Always enable mitigation in the
6235 interest in being mitigated too.
6239 is in page units and it defines how many pages prior
6241 growing up) the main stack are reserved for no other
6253 stacktrace_filter=[function-list]
6255 will trace at boot up. function-list is a comma-separated
6257 time by the stack_trace_filter file in the debugfs
6263 Set the STI (builtin display/keyboard on the HP-PARISC
6265 as the initial boot-console.
6266 See also comment in drivers/video/console/sticore.c.
6269 See comment in drivers/video/console/sticore.c.
6284 Limits the number of kernel HPT entries in the hash
6297 originate from a privileged port (i.e. a port in the
6308 process in parallel from a single connection.
6309 The default value is 0 (no limit).
6325 to global on non-NUMA machines)
6336 suspend.pm_test_delay=
6337 [SUSPEND]
6338 Sets the number of seconds to remain in a suspend test
6348 swiotlb= [ARM,IA-64,PPC,MIPS,X86]
6350 <int> -- Number of I/O TLB slabs
6351 <int> -- Second integer after comma. Number of swiotlb
6354 force -- force using of bounce buffers even if they
6356 noforce -- Never use bounce buffers (for debugging)
6365 are reported in the kernel log. Sysctls registered
6371 Ignore sysrq setting - this boot parameter will
6379 cache size. See Documentation/networking/ip-sysctl.rst
6384 test_suspend= [SUSPEND]
6386 Specify "mem" (for Suspend-to-RAM) or "standby" (for
6387 standby suspend) or "freeze" (for suspend type freeze)
6391 wakeup-capable RTC alarm.
6397 -1: disable all active trip points in all thermal zones
6401 -1: disable all critical trip points in all thermal zones
6408 -1: disable all passive trip points
6414 <deci-seconds>: poll all this frequency
6415 0: no polling (default)
6436 Prevent the CPU-hotplug component of torturing
6440 Dump the ftrace buffer at torture-test shutdown,
6441 even if there were no errors. This can be a
6444 with rotating-rust storage.
6449 disables verbose-printk() sleeping.
6452 Duration of each verbose-printk() sleep in jiffies.
6458 Specify that at suspend time, the tpm driver
6470 https://trustedcomputinggroup.org/resource/pc-client-platform-tpm-profile-ptp-specification/
6482 tracepoint_printk kernel cmdline option has no effect.
6508 local - Use the per CPU time stamp counter
6511 in sync between CPUs.
6512 global - Event time stamps are synchronize across
6515 counter - Simple counting of events (1, 2, ..)
6519 uptime - Use jiffies as the time stamp.
6520 perf - Use the same clock that perf uses.
6521 mono - Use ktime_get_mono_fast_ns() for time stamps.
6522 mono_raw - Use ktime_get_raw_fast_ns() for time
6524 boot - Use ktime_get_boot_fast_ns() for time stamps.
6528 trace_event=[event-list]
6529 [FTRACE] Set and start specified trace events in order
6530 to facilitate early boot debugging. The event-list is a
6531 comma-separated list of trace events to enable. See
6534 trace_instance=[instance-info]
6535 [FTRACE] Create a ring buffer instance early in boot up.
6536 This will be listed in:
6554 trace_options=[option-list]
6556 The option-list is a comma delimited list of options
6570 trace_trigger=[trigger-list]
6586 See also "Event triggers" in Documentation/trace/events.rst
6593 file located in /sys/kernel/tracing/
6607 See Documentation/admin-guide/mm/transhuge.rst
6615 - "tpm"
6616 - "tee"
6617 - "caam"
6627 - "kernel"
6628 - the same value as trusted.source: "tpm" or "tee"
6629 - "default"
6630 If not specified, "default" is used. In this case,
6638 high-resolution timer mode on older hardware, and in
6648 in situations with strict latency requirements (where
6669 Extensions (TSX) feature in Intel processors that
6674 on - Enable TSX on the system. Although there are
6677 several previous speculation-related CVEs, and
6681 off - Disable TSX on the system. (Note that this
6689 auto - Disable TSX if X86_BUG_TAA is present,
6694 See Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
6700 Similar to Micro-architectural Data Sampling (MDS)
6707 In vulnerable processors, the speculatively forwarded
6708 data can be used in a cache side channel attack, to
6715 full - Enable TAA mitigation on vulnerable CPUs
6718 full,nosmt - Enable TAA mitigation and disable SMT on
6721 vulnerable to cross-thread TAA attacks.
6722 off - Unconditionally disable TAA mitigation
6724 On MDS-affected machines, tsx_async_abort=off can be
6726 are mitigated with the same mechanism so in order to disable
6736 Documentation/admin-guide/hw-vuln/tsx_async_abort.rst
6742 See also Documentation/input/devices/joystick-parport.rst
6744 udbg-immortal [PPC] When debugging early kernel crashes that
6750 Set number of hash buckets for UDP/UDP-Lite connections
6752 uhci-hcd.ignore_oc=
6754 Some badly-designed motherboards generate lots of
6763 unwind_debug [X86-64]
6771 (default -1 = authorized (same as 1),
6776 [USB] The autosuspend time delay (in seconds) used
6777 for newly-detected USB devices (default 2). This
6786 [USB] Maximum number of bytes to snoop in each URB
6797 [USB] Memory limit (in MB) for buffers allocated by
6805 [USB] Specifies timeout for the initial 64-byte
6806 USB_REQ_GET_DESCRIPTOR request in milliseconds
6812 [USB] A list of quirk entries to augment the built-in
6815 VendorID:ProductID:Flags. The IDs are 4-digit hex
6817 will change the built-in quirk; setting it if it is
6822 a 255-byte read);
6826 Set-Interface requests);
6842 require the interval in microframes (1
6845 (bInterval-1).
6849 variable used in the calculation);
6863 to be disconnected before suspend to
6883 usb-storage.delay_use=
6884 [UMS] The delay in seconds before a new device is
6887 usb-storage.quirks=
6889 override the built-in unusual_devs list. List
6892 and Product ID values (4-digit hex numbers) and
6894 to a common usb-storage quirk flag as follows:
6936 medium is write-protected).
6938 even if the device claims no cache,
6945 1 - undefined instruction events
6946 2 - system calls
6947 4 - invalid data aborts
6948 8 - SIGSEGV faults
6949 16 - SIGBUS faults
6955 nohigh = do not allocate PTE pages in
6965 vdso32= [X86] Control the 32-bit vDSO
6966 vdso32=1: enable 32-bit VDSO
6967 vdso32=0 or vdso32=2: disable 32-bit VDSO
6977 dl_main: Assertion `(void *) ph->p_vaddr == _rtld_local._dl_sysinfo_dso' failed!
6979 vector= [IA-64,SMP]
7011 vga= [BOOT,X86-32] Select a particular video mode
7013 Documentation/admin-guide/svga.rst.
7028 - Disable all of the above options
7049 vsyscall= [X86-64]
7052 code). Most statically-linked binaries and older
7076 see VGA-softcursor.txt. Default: 2 = underline.
7081 This is a 16-member array composed of values
7082 ranging from 0-255.
7087 This is a 16-member array composed of values
7088 ranging from 0-255.
7093 This is a 16-member array composed of values
7094 ranging from 0-255.
7099 Set system-wide default UTF-8 mode for all tty's.
7100 Default is 1, i.e. UTF-8 mode is enabled for all
7105 Format=<-1|0|1>
7106 Set system-wide default for whether a cursor
7107 is shown on new VTs. Default is -1,
7112 vt.italic= [VT] Default color for italic text; 0-15.
7115 vt.underline= [VT] Default color for underlined text; 0-15.
7119 see Documentation/watchdog/watchdog-parameters.rst
7120 or other driver-specific files in the
7126 threshold in seconds. The soft lockup detector
7133 to use in unbound workqueues.
7134 Format: <cpu-list>
7143 duration in seconds. The default value is 30 and
7148 Per-cpu work items which run for longer than this
7151 them from noticeably delaying other per-cpu work
7160 Per-cpu workqueues are generally preferred because
7162 locality; unfortunately, per-cpu workqueues tend to
7165 Enabling this makes the per-cpu workqueues which
7178 information, see the Affinity Scopes section in
7179 Documentation/core-api/workqueue.rst.
7189 on the local CPU. This guarantee is no longer true
7192 forces round-robin CPU selection to flush out
7200 on - Enable writecombine, use WUC for ioremap_wc()
7201 off - Disable writecombine, use SUC for ioremap_wc()
7203 x2apic_phys [X86-64,APIC] Use x2apic physical mode instead of
7207 xen_512gb_limit [KNL,X86-64,XEN]
7217 ide-disks -- unplug primary master IDE devices
7218 aux-ide-disks -- unplug non-primary-master IDE devices
7219 nics -- unplug network devices
7220 all -- unplug all emulated devices (NICs and IDE disks)
7221 unnecessary -- unplugging emulated devices is
7224 never -- do not unplug even if version check succeeds
7232 Select whether to always use non-faulting (safe) MSR
7243 run as generic HVM guest with no PV drivers.
7257 xen_timer_slop= [X86-64,XEN]
7258 Set the timer slop (in nanoseconds) for the virtual Xen
7265 The time (in seconds) to wait before giving up to boot
7266 in case initial ballooning fails to free enough memory.
7272 How long to delay EOI handling in case of event
7282 preferred over the 2-level event handling, as it is
7299 xive.store-eoi=off [PPC]
7305 xhci-hcd.quirks [USB,KNL]
7308 consulted in header drivers/usb/host/xhci.h.