2 mainmenu "Linux Kernel Configuration for x86"
6 bool "64-bit kernel" if ARCH = "x86"
7 default ARCH = "x86_64"
9 Say yes to build a 64-bit kernel - formerly known as x86_64
10 Say no to build a 32-bit kernel - formerly known as i386
24 select HAVE_KRETPROBES
25 select HAVE_KVM if ((X86_32 && !X86_VOYAGER && !X86_VISWS && !X86_NUMAQ) || X86_64)
29 config GENERIC_LOCKBREAK
35 config GENERIC_CMOS_UPDATE
38 config CLOCKSOURCE_WATCHDOG
41 config GENERIC_CLOCKEVENTS
44 config GENERIC_CLOCKEVENTS_BROADCAST
46 depends on X86_64 || (X86_32 && X86_LOCAL_APIC)
48 config LOCKDEP_SUPPORT
51 config STACKTRACE_SUPPORT
54 config HAVE_LATENCYTOP_SUPPORT
57 config FAST_CMPXCHG_LOCAL
70 config GENERIC_ISA_DMA
80 config GENERIC_HWEIGHT
86 config ARCH_MAY_HAVE_PC_FDC
92 config RWSEM_GENERIC_SPINLOCK
95 config RWSEM_XCHGADD_ALGORITHM
98 config ARCH_HAS_ILOG2_U32
101 config ARCH_HAS_ILOG2_U64
104 config ARCH_HAS_CPU_IDLE_WAIT
107 config GENERIC_CALIBRATE_DELAY
110 config GENERIC_TIME_VSYSCALL
114 config ARCH_HAS_CPU_RELAX
117 config HAVE_SETUP_PER_CPU_AREA
118 def_bool X86_64 || (X86_SMP && !X86_VOYAGER)
120 config ARCH_HIBERNATION_POSSIBLE
122 depends on !SMP || !X86_VOYAGER
124 config ARCH_SUSPEND_POSSIBLE
126 depends on !X86_VOYAGER
132 config ARCH_POPULATES_NODE_MAP
139 config ARCH_SUPPORTS_AOUT
142 # Use the generic interrupt handling code in kernel/irq/:
143 config GENERIC_HARDIRQS
147 config GENERIC_IRQ_PROBE
151 config GENERIC_PENDING_IRQ
153 depends on GENERIC_HARDIRQS && SMP
158 depends on SMP && ((X86_32 && !X86_VOYAGER) || X86_64)
163 depends on X86_32 && SMP
167 depends on X86_64 && SMP
172 depends on (X86_32 && !(X86_VISWS || X86_VOYAGER)) || X86_64
175 config X86_BIOS_REBOOT
177 depends on X86_32 && !(X86_VISWS || X86_VOYAGER)
180 config X86_TRAMPOLINE
182 depends on X86_SMP || (X86_VOYAGER && SMP) || (64BIT && ACPI_SLEEP)
187 source "init/Kconfig"
189 menu "Processor type and features"
191 source "kernel/time/Kconfig"
194 bool "Symmetric multi-processing support"
196 This enables support for systems with more than one CPU. If you have
197 a system with only one CPU, like most personal computers, say N. If
198 you have a system with more than one CPU, say Y.
200 If you say N here, the kernel will run on single and multiprocessor
201 machines, but will use only one CPU of a multiprocessor machine. If
202 you say Y here, the kernel will run on many, but not all,
203 singleprocessor machines. On a singleprocessor machine, the kernel
204 will run faster if you say N here.
206 Note that if you say Y here and choose architecture "586" or
207 "Pentium" under "Processor family", the kernel will not work on 486
208 architectures. Similarly, multiprocessor kernels for the "PPro"
209 architecture may not work on all Pentium based boards.
211 People using multiprocessor machines who say Y here should also say
212 Y to "Enhanced Real Time Clock Support", below. The "Advanced Power
213 Management" code will be disabled if you say Y here.
215 See also <file:Documentation/i386/IO-APIC.txt>,
216 <file:Documentation/nmi_watchdog.txt> and the SMP-HOWTO available at
217 <http://www.tldp.org/docs.html#howto>.
219 If you don't know what to do here, say N.
222 prompt "Subarchitecture Type"
228 Choose this option if your computer is a standard PC or compatible.
234 Select this for an AMD Elan processor.
236 Do not use this option for K6/Athlon/Opteron processors!
238 If unsure, choose "PC-compatible" instead.
242 depends on X86_32 && (SMP || BROKEN)
244 Voyager is an MCA-based 32-way capable SMP architecture proprietary
245 to NCR Corp. Machine classes 345x/35xx/4100/51xx are Voyager-based.
249 If you do not specifically know you have a Voyager based machine,
250 say N here, otherwise the kernel you build will not be bootable.
253 bool "NUMAQ (IBM/Sequent)"
254 depends on SMP && X86_32
257 This option is used for getting Linux to run on a (IBM/Sequent) NUMA
258 multiquad box. This changes the way that processors are bootstrapped,
259 and uses Clustered Logical APIC addressing mode instead of Flat Logical.
260 You will need a new lynxer.elf file to flash your firmware with - send
261 email to <Martin.Bligh@us.ibm.com>.
264 bool "Summit/EXA (IBM x440)"
265 depends on X86_32 && SMP
267 This option is needed for IBM systems that use the Summit/EXA chipset.
268 In particular, it is needed for the x440.
270 If you don't have one of these computers, you should say N here.
271 If you want to build a NUMA kernel, you must select ACPI.
274 bool "Support for other sub-arch SMP systems with more than 8 CPUs"
275 depends on X86_32 && SMP
277 This option is needed for the systems that have more than 8 CPUs
278 and if the system is not of any sub-arch type above.
280 If you don't have such a system, you should say N here.
283 bool "SGI 320/540 (Visual Workstation)"
286 The SGI Visual Workstation series is an IA32-based workstation
287 based on SGI systems chips with some legacy PC hardware attached.
289 Say Y here to create a kernel to run on the SGI 320 or 540.
291 A kernel compiled for the Visual Workstation will not run on PCs
292 and vice versa. See <file:Documentation/sgi-visws.txt> for details.
294 config X86_GENERICARCH
295 bool "Generic architecture (Summit, bigsmp, ES7000, default)"
298 This option compiles in the Summit, bigsmp, ES7000, default subarchitectures.
299 It is intended for a generic binary kernel.
300 If you want a NUMA kernel, select ACPI. We need SRAT for NUMA.
303 bool "Support for Unisys ES7000 IA32 series"
304 depends on X86_32 && SMP
306 Support for Unisys ES7000 systems. Say 'Y' here if this kernel is
307 supposed to run on an IA32-based Unisys ES7000 system.
308 Only choose this option if you have such a system, otherwise you
312 bool "RDC R-321x SoC"
315 select X86_REBOOTFIXUPS
320 This option is needed for RDC R-321x system-on-chip, also known
322 If you don't have one of these chips, you should say N here.
325 bool "Support for ScaleMP vSMP"
329 Support for ScaleMP vSMP systems. Say 'Y' here if this kernel is
330 supposed to run on these EM64T-based machines. Only choose this option
331 if you have one of these machines.
335 config SCHED_NO_NO_OMIT_FRAME_POINTER
337 prompt "Single-depth WCHAN output"
340 Calculate simpler /proc/<PID>/wchan values. If this option
341 is disabled then wchan values will recurse back to the
342 caller function. This provides more accurate wchan values,
343 at the expense of slightly more scheduling overhead.
345 If in doubt, say "Y".
347 menuconfig PARAVIRT_GUEST
348 bool "Paravirtualized guest support"
350 Say Y here to get to see options related to running Linux under
351 various hypervisors. This option alone does not add any kernel code.
353 If you say N, all options in this submenu will be skipped and disabled.
357 source "arch/x86/xen/Kconfig"
360 bool "VMI Guest support"
363 depends on !(X86_VISWS || X86_VOYAGER)
365 VMI provides a paravirtualized interface to the VMware ESX server
366 (it could be used by other hypervisors in theory too, but is not
367 at the moment), by linking the kernel to a GPL-ed ROM module
368 provided by the hypervisor.
370 source "arch/x86/lguest/Kconfig"
373 bool "Enable paravirtualization code"
374 depends on !(X86_VISWS || X86_VOYAGER)
376 This changes the kernel so it can modify itself when it is run
377 under a hypervisor, potentially improving performance significantly
378 over full virtualization. However, when run without a hypervisor
379 the kernel is theoretically slower and slightly larger.
383 config MEMTEST_BOOTPARAM
384 bool "Memtest boot parameter"
388 This option adds a kernel parameter 'memtest', which allows memtest
389 to be disabled at boot. If this option is selected, memtest
390 functionality can be disabled with memtest=0 on the kernel
391 command line. The purpose of this option is to allow a single
392 kernel image to be distributed with memtest built in, but not
395 If you are unsure how to answer this question, answer Y.
397 config MEMTEST_BOOTPARAM_VALUE
398 int "Memtest boot parameter default value (0-4)"
399 depends on MEMTEST_BOOTPARAM
403 This option sets the default value for the kernel parameter
404 'memtest', which allows memtest to be disabled at boot. If this
405 option is set to 0 (zero), the memtest kernel parameter will
406 default to 0, disabling memtest at bootup. If this option is
407 set to 4, the memtest kernel parameter will default to 4,
408 enabling memtest at bootup, and use that as pattern number.
410 If you are unsure how to answer this question, answer 0.
414 depends on X86_32 && ACPI && NUMA && (X86_SUMMIT || X86_GENERICARCH)
417 config HAVE_ARCH_PARSE_SRAT
421 config X86_SUMMIT_NUMA
423 depends on X86_32 && NUMA && (X86_SUMMIT || X86_GENERICARCH)
425 config X86_CYCLONE_TIMER
427 depends on X86_32 && X86_SUMMIT || X86_GENERICARCH
429 config ES7000_CLUSTERED_APIC
431 depends on SMP && X86_ES7000 && MPENTIUMIII
433 source "arch/x86/Kconfig.cpu"
437 prompt "HPET Timer Support" if X86_32
439 Use the IA-PC HPET (High Precision Event Timer) to manage
440 time in preference to the PIT and RTC, if a HPET is
442 HPET is the next generation timer replacing legacy 8254s.
443 The HPET provides a stable time base on SMP
444 systems, unlike the TSC, but it is more expensive to access,
445 as it is off-chip. You can find the HPET spec at
446 <http://www.intel.com/hardwaredesign/hpetspec.htm>.
448 You can safely choose Y here. However, HPET will only be
449 activated if the platform and the BIOS support this feature.
450 Otherwise the 8254 will be used for timing services.
452 Choose N to continue using the legacy 8254 timer.
454 config HPET_EMULATE_RTC
456 depends on HPET_TIMER && (RTC=y || RTC=m || RTC_DRV_CMOS=m || RTC_DRV_CMOS=y)
458 # Mark as embedded because too many people got it wrong.
459 # The code disables itself when not needed.
461 bool "GART IOMMU support" if EMBEDDED
465 depends on X86_64 && PCI
467 Support for full DMA access of devices with 32bit memory access only
468 on systems with more than 3GB. This is usually needed for USB,
469 sound, many IDE/SATA chipsets and some other devices.
470 Provides a driver for the AMD Athlon64/Opteron/Turion/Sempron GART
471 based hardware IOMMU and a software bounce buffer based IOMMU used
472 on Intel systems and as fallback.
473 The code is only active when needed (enough memory and limited
474 device) unless CONFIG_IOMMU_DEBUG or iommu=force is specified
478 bool "IBM Calgary IOMMU support"
480 depends on X86_64 && PCI && EXPERIMENTAL
482 Support for hardware IOMMUs in IBM's xSeries x366 and x460
483 systems. Needed to run systems with more than 3GB of memory
484 properly with 32-bit PCI devices that do not support DAC
485 (Double Address Cycle). Calgary also supports bus level
486 isolation, where all DMAs pass through the IOMMU. This
487 prevents them from going anywhere except their intended
488 destination. This catches hard-to-find kernel bugs and
489 mis-behaving drivers and devices that do not use the DMA-API
490 properly to set up their DMA buffers. The IOMMU can be
491 turned off at boot time with the iommu=off parameter.
492 Normally the kernel will make the right choice by itself.
495 config CALGARY_IOMMU_ENABLED_BY_DEFAULT
497 prompt "Should Calgary be enabled by default?"
498 depends on CALGARY_IOMMU
500 Should Calgary be enabled by default? if you choose 'y', Calgary
501 will be used (if it exists). If you choose 'n', Calgary will not be
502 used even if it exists. If you choose 'n' and would like to use
503 Calgary anyway, pass 'iommu=calgary' on the kernel command line.
507 def_bool (CALGARY_IOMMU || GART_IOMMU)
509 # need this always selected by IOMMU for the VIA workaround
513 Support for software bounce buffers used on x86-64 systems
514 which don't have a hardware IOMMU (e.g. the current generation
515 of Intel's x86-64 CPUs). Using this PCI devices which can only
516 access 32-bits of memory can be used on systems with more than
517 3 GB of memory. If unsure, say Y.
521 int "Maximum number of CPUs (2-255)"
524 default "32" if X86_NUMAQ || X86_SUMMIT || X86_BIGSMP || X86_ES7000
527 This allows you to specify the maximum number of CPUs which this
528 kernel will support. The maximum supported value is 255 and the
529 minimum value which makes sense is 2.
531 This is purely to save memory - each supported CPU adds
532 approximately eight kilobytes to the kernel image.
535 bool "SMT (Hyperthreading) scheduler support"
538 SMT scheduler support improves the CPU scheduler's decision making
539 when dealing with Intel Pentium 4 chips with HyperThreading at a
540 cost of slightly increased overhead in some places. If unsure say
545 prompt "Multi-core scheduler support"
548 Multi-core scheduler support improves the CPU scheduler's decision
549 making when dealing with multi-core CPU chips at a cost of slightly
550 increased overhead in some places. If unsure say N here.
552 source "kernel/Kconfig.preempt"
555 bool "Local APIC support on uniprocessors"
556 depends on X86_32 && !SMP && !(X86_VISWS || X86_VOYAGER || X86_GENERICARCH)
558 A local APIC (Advanced Programmable Interrupt Controller) is an
559 integrated interrupt controller in the CPU. If you have a single-CPU
560 system which has a processor with a local APIC, you can say Y here to
561 enable and use it. If you say Y here even though your machine doesn't
562 have a local APIC, then the kernel will still run with no slowdown at
563 all. The local APIC supports CPU-generated self-interrupts (timer,
564 performance counters), and the NMI watchdog which detects hard
568 bool "IO-APIC support on uniprocessors"
569 depends on X86_UP_APIC
571 An IO-APIC (I/O Advanced Programmable Interrupt Controller) is an
572 SMP-capable replacement for PC-style interrupt controllers. Most
573 SMP systems and many recent uniprocessor systems have one.
575 If you have a single-CPU system with an IO-APIC, you can say Y here
576 to use it. If you say Y here even though your machine doesn't have
577 an IO-APIC, then the kernel will still run with no slowdown at all.
579 config X86_LOCAL_APIC
581 depends on X86_64 || (X86_32 && (X86_UP_APIC || ((X86_VISWS || SMP) && !X86_VOYAGER) || X86_GENERICARCH))
585 depends on X86_64 || (X86_32 && (X86_UP_IOAPIC || (SMP && !(X86_VISWS || X86_VOYAGER)) || X86_GENERICARCH))
587 config X86_VISWS_APIC
589 depends on X86_32 && X86_VISWS
592 bool "Machine Check Exception"
593 depends on !X86_VOYAGER
595 Machine Check Exception support allows the processor to notify the
596 kernel if it detects a problem (e.g. overheating, component failure).
597 The action the kernel takes depends on the severity of the problem,
598 ranging from a warning message on the console, to halting the machine.
599 Your processor must be a Pentium or newer to support this - check the
600 flags in /proc/cpuinfo for mce. Note that some older Pentium systems
601 have a design flaw which leads to false MCE events - hence MCE is
602 disabled on all P5 processors, unless explicitly enabled with "mce"
603 as a boot argument. Similarly, if MCE is built in and creates a
604 problem on some new non-standard machine, you can boot with "nomce"
605 to disable it. MCE support simply ignores non-MCE processors like
606 the 386 and 486, so nearly everyone can say Y here.
610 prompt "Intel MCE features"
611 depends on X86_64 && X86_MCE && X86_LOCAL_APIC
613 Additional support for intel specific MCE features such as
618 prompt "AMD MCE features"
619 depends on X86_64 && X86_MCE && X86_LOCAL_APIC
621 Additional support for AMD specific MCE features such as
622 the DRAM Error Threshold.
624 config X86_MCE_NONFATAL
625 tristate "Check for non-fatal errors on AMD Athlon/Duron / Intel Pentium 4"
626 depends on X86_32 && X86_MCE
628 Enabling this feature starts a timer that triggers every 5 seconds which
629 will look at the machine check registers to see if anything happened.
630 Non-fatal problems automatically get corrected (but still logged).
631 Disable this if you don't want to see these messages.
632 Seeing the messages this option prints out may be indicative of dying
633 or out-of-spec (ie, overclocked) hardware.
634 This option only does something on certain CPUs.
635 (AMD Athlon/Duron and Intel Pentium 4)
637 config X86_MCE_P4THERMAL
638 bool "check for P4 thermal throttling interrupt."
639 depends on X86_32 && X86_MCE && (X86_UP_APIC || SMP) && !X86_VISWS
641 Enabling this feature will cause a message to be printed when the P4
642 enters thermal throttling.
645 bool "Enable VM86 support" if EMBEDDED
649 This option is required by programs like DOSEMU to run 16-bit legacy
650 code on X86 processors. It also may be needed by software like
651 XFree86 to initialize some video cards via BIOS. Disabling this
652 option saves about 6k.
655 tristate "Toshiba Laptop support"
658 This adds a driver to safely access the System Management Mode of
659 the CPU on Toshiba portables with a genuine Toshiba BIOS. It does
660 not work on models with a Phoenix BIOS. The System Management Mode
661 is used to set the BIOS and power saving options on Toshiba portables.
663 For information on utilities to make use of this driver see the
664 Toshiba Linux utilities web site at:
665 <http://www.buzzard.org.uk/toshiba/>.
667 Say Y if you intend to run this kernel on a Toshiba portable.
671 tristate "Dell laptop support"
673 This adds a driver to safely access the System Management Mode
674 of the CPU on the Dell Inspiron 8000. The System Management Mode
675 is used to read cpu temperature and cooling fan status and to
676 control the fans on the I8K portables.
678 This driver has been tested only on the Inspiron 8000 but it may
679 also work with other Dell laptops. You can force loading on other
680 models by passing the parameter `force=1' to the module. Use at
683 For information on utilities to make use of this driver see the
684 I8K Linux utilities web site at:
685 <http://people.debian.org/~dz/i8k/>
687 Say Y if you intend to run this kernel on a Dell Inspiron 8000.
690 config X86_REBOOTFIXUPS
692 prompt "Enable X86 board specific fixups for reboot"
693 depends on X86_32 && X86
695 This enables chipset and/or board specific fixups to be done
696 in order to get reboot to work correctly. This is only needed on
697 some combinations of hardware and BIOS. The symptom, for which
698 this config is intended, is when reboot ends with a stalled/hung
701 Currently, the only fixup is for the Geode machines using
702 CS5530A and CS5536 chipsets and the RDC R-321x SoC.
704 Say Y if you want to enable the fixup. Currently, it's safe to
705 enable this option even if you don't need it.
709 tristate "/dev/cpu/microcode - Intel IA32 CPU microcode support"
712 If you say Y here, you will be able to update the microcode on
713 Intel processors in the IA32 family, e.g. Pentium Pro, Pentium II,
714 Pentium III, Pentium 4, Xeon etc. You will obviously need the
715 actual microcode binary data itself which is not shipped with the
718 For latest news and information on obtaining all the required
719 ingredients for this driver, check:
720 <http://www.urbanmyth.org/microcode/>.
722 To compile this driver as a module, choose M here: the
723 module will be called microcode.
725 config MICROCODE_OLD_INTERFACE
730 tristate "/dev/cpu/*/msr - Model-specific register support"
732 This device gives privileged processes access to the x86
733 Model-Specific Registers (MSRs). It is a character device with
734 major 202 and minors 0 to 31 for /dev/cpu/0/msr to /dev/cpu/31/msr.
735 MSR accesses are directed to a specific CPU on multi-processor
739 tristate "/dev/cpu/*/cpuid - CPU information support"
741 This device gives processes access to the x86 CPUID instruction to
742 be executed on a specific processor. It is a character device
743 with major 203 and minors 0 to 31 for /dev/cpu/0/cpuid to
747 prompt "High Memory Support"
748 default HIGHMEM4G if !X86_NUMAQ
749 default HIGHMEM64G if X86_NUMAQ
754 depends on !X86_NUMAQ
756 Linux can use up to 64 Gigabytes of physical memory on x86 systems.
757 However, the address space of 32-bit x86 processors is only 4
758 Gigabytes large. That means that, if you have a large amount of
759 physical memory, not all of it can be "permanently mapped" by the
760 kernel. The physical memory that's not permanently mapped is called
763 If you are compiling a kernel which will never run on a machine with
764 more than 1 Gigabyte total physical RAM, answer "off" here (default
765 choice and suitable for most users). This will result in a "3GB/1GB"
766 split: 3GB are mapped so that each process sees a 3GB virtual memory
767 space and the remaining part of the 4GB virtual memory space is used
768 by the kernel to permanently map as much physical memory as
771 If the machine has between 1 and 4 Gigabytes physical RAM, then
774 If more than 4 Gigabytes is used then answer "64GB" here. This
775 selection turns Intel PAE (Physical Address Extension) mode on.
776 PAE implements 3-level paging on IA32 processors. PAE is fully
777 supported by Linux, PAE mode is implemented on all recent Intel
778 processors (Pentium Pro and better). NOTE: If you say "64GB" here,
779 then the kernel will not boot on CPUs that don't support PAE!
781 The actual amount of total physical memory will either be
782 auto detected or can be forced by using a kernel command line option
783 such as "mem=256M". (Try "man bootparam" or see the documentation of
784 your boot loader (lilo or loadlin) about how to pass options to the
785 kernel at boot time.)
787 If unsure, say "off".
791 depends on !X86_NUMAQ
793 Select this if you have a 32-bit processor and between 1 and 4
794 gigabytes of physical RAM.
798 depends on !M386 && !M486
801 Select this if you have a 32-bit processor and more than 4
802 gigabytes of physical RAM.
807 depends on EXPERIMENTAL
808 prompt "Memory split" if EMBEDDED
812 Select the desired split between kernel and user memory.
814 If the address range available to the kernel is less than the
815 physical memory installed, the remaining memory will be available
816 as "high memory". Accessing high memory is a little more costly
817 than low memory, as it needs to be mapped into the kernel first.
818 Note that increasing the kernel address space limits the range
819 available to user programs, making the address space there
820 tighter. Selecting anything other than the default 3G/1G split
821 will also likely make your kernel incompatible with binary-only
824 If you are not absolutely sure what you are doing, leave this
828 bool "3G/1G user/kernel split"
829 config VMSPLIT_3G_OPT
831 bool "3G/1G user/kernel split (for full 1G low memory)"
833 bool "2G/2G user/kernel split"
834 config VMSPLIT_2G_OPT
836 bool "2G/2G user/kernel split (for full 2G low memory)"
838 bool "1G/3G user/kernel split"
843 default 0xB0000000 if VMSPLIT_3G_OPT
844 default 0x80000000 if VMSPLIT_2G
845 default 0x78000000 if VMSPLIT_2G_OPT
846 default 0x40000000 if VMSPLIT_1G
852 depends on X86_32 && (HIGHMEM64G || HIGHMEM4G)
856 prompt "PAE (Physical Address Extension) Support"
857 depends on X86_32 && !HIGHMEM4G
858 select RESOURCES_64BIT
860 PAE is required for NX support, and furthermore enables
861 larger swapspace support for non-overcommit purposes. It
862 has the cost of more pagetable lookup overhead, and also
863 consumes more pagetable space per process.
865 # Common NUMA Features
867 bool "Numa Memory Allocation and Scheduler Support (EXPERIMENTAL)"
869 depends on X86_64 || (X86_32 && HIGHMEM64G && (X86_NUMAQ || (X86_SUMMIT || X86_GENERICARCH) && ACPI) && EXPERIMENTAL)
871 default y if (X86_NUMAQ || X86_SUMMIT)
873 Enable NUMA (Non Uniform Memory Access) support.
874 The kernel will try to allocate memory used by a CPU on the
875 local memory controller of the CPU and add some more
876 NUMA awareness to the kernel.
878 For i386 this is currently highly experimental and should be only
879 used for kernel development. It might also cause boot failures.
880 For x86_64 this is recommended on all multiprocessor Opteron systems.
881 If the system is EM64T, you should say N unless your system is
884 comment "NUMA (Summit) requires SMP, 64GB highmem support, ACPI"
885 depends on X86_32 && X86_SUMMIT && (!HIGHMEM64G || !ACPI)
889 prompt "Old style AMD Opteron NUMA detection"
890 depends on X86_64 && NUMA && PCI
892 Enable K8 NUMA node topology detection. You should say Y here if
893 you have a multi processor AMD K8 system. This uses an old
894 method to read the NUMA configuration directly from the builtin
895 Northbridge of Opteron. It is recommended to use X86_64_ACPI_NUMA
896 instead, which also takes priority if both are compiled in.
898 config X86_64_ACPI_NUMA
900 prompt "ACPI NUMA detection"
901 depends on X86_64 && NUMA && ACPI && PCI
904 Enable ACPI SRAT based node topology detection.
907 bool "NUMA emulation"
908 depends on X86_64 && NUMA
910 Enable NUMA emulation. A flat machine will be split
911 into virtual nodes when booted with "numa=fake=N", where N is the
912 number of nodes. This is only useful for debugging.
915 int "Max num nodes shift(1-15)"
917 default "6" if X86_64
918 default "4" if X86_NUMAQ
920 depends on NEED_MULTIPLE_NODES
922 config HAVE_ARCH_BOOTMEM_NODE
924 depends on X86_32 && NUMA
926 config ARCH_HAVE_MEMORY_PRESENT
928 depends on X86_32 && DISCONTIGMEM
930 config NEED_NODE_MEMMAP_SIZE
932 depends on X86_32 && (DISCONTIGMEM || SPARSEMEM)
934 config HAVE_ARCH_ALLOC_REMAP
936 depends on X86_32 && NUMA
938 config ARCH_FLATMEM_ENABLE
940 depends on X86_32 && ARCH_SELECT_MEMORY_MODEL && X86_PC && !NUMA
942 config ARCH_DISCONTIGMEM_ENABLE
944 depends on NUMA && X86_32
946 config ARCH_DISCONTIGMEM_DEFAULT
948 depends on NUMA && X86_32
950 config ARCH_SPARSEMEM_DEFAULT
954 config ARCH_SPARSEMEM_ENABLE
956 depends on X86_64 || NUMA || (EXPERIMENTAL && X86_PC)
957 select SPARSEMEM_STATIC if X86_32
958 select SPARSEMEM_VMEMMAP_ENABLE if X86_64
960 config ARCH_SELECT_MEMORY_MODEL
962 depends on ARCH_SPARSEMEM_ENABLE
964 config ARCH_MEMORY_PROBE
966 depends on MEMORY_HOTPLUG
971 bool "Allocate 3rd-level pagetables from highmem"
972 depends on X86_32 && (HIGHMEM4G || HIGHMEM64G)
974 The VM uses one page table entry for each page of physical memory.
975 For systems with a lot of RAM, this can be wasteful of precious
976 low memory. Setting this option will put user-space page table
977 entries in high memory.
979 config MATH_EMULATION
981 prompt "Math emulation" if X86_32
983 Linux can emulate a math coprocessor (used for floating point
984 operations) if you don't have one. 486DX and Pentium processors have
985 a math coprocessor built in, 486SX and 386 do not, unless you added
986 a 487DX or 387, respectively. (The messages during boot time can
987 give you some hints here ["man dmesg"].) Everyone needs either a
988 coprocessor or this emulation.
990 If you don't have a math coprocessor, you need to say Y here; if you
991 say Y here even though you have a coprocessor, the coprocessor will
992 be used nevertheless. (This behavior can be changed with the kernel
993 command line option "no387", which comes handy if your coprocessor
994 is broken. Try "man bootparam" or see the documentation of your boot
995 loader (lilo or loadlin) about how to pass options to the kernel at
996 boot time.) This means that it is a good idea to say Y here if you
997 intend to use this kernel on different machines.
999 More information about the internals of the Linux math coprocessor
1000 emulation can be found in <file:arch/x86/math-emu/README>.
1002 If you are not sure, say Y; apart from resulting in a 66 KB bigger
1003 kernel, it won't hurt.
1006 bool "MTRR (Memory Type Range Register) support"
1008 On Intel P6 family processors (Pentium Pro, Pentium II and later)
1009 the Memory Type Range Registers (MTRRs) may be used to control
1010 processor access to memory ranges. This is most useful if you have
1011 a video (VGA) card on a PCI or AGP bus. Enabling write-combining
1012 allows bus write transfers to be combined into a larger transfer
1013 before bursting over the PCI/AGP bus. This can increase performance
1014 of image write operations 2.5 times or more. Saying Y here creates a
1015 /proc/mtrr file which may be used to manipulate your processor's
1016 MTRRs. Typically the X server should use this.
1018 This code has a reasonably generic interface so that similar
1019 control registers on other processors can be easily supported
1022 The Cyrix 6x86, 6x86MX and M II processors have Address Range
1023 Registers (ARRs) which provide a similar functionality to MTRRs. For
1024 these, the ARRs are used to emulate the MTRRs.
1025 The AMD K6-2 (stepping 8 and above) and K6-3 processors have two
1026 MTRRs. The Centaur C6 (WinChip) has 8 MCRs, allowing
1027 write-combining. All of these processors are supported by this code
1028 and it makes sense to say Y here if you have one of them.
1030 Saying Y here also fixes a problem with buggy SMP BIOSes which only
1031 set the MTRRs for the boot CPU and not for the secondary CPUs. This
1032 can lead to all sorts of problems, so it's good to say Y here.
1034 You can safely say Y even if your machine doesn't have MTRRs, you'll
1035 just add about 9 KB to your kernel.
1037 See <file:Documentation/mtrr.txt> for more information.
1041 prompt "x86 PAT support"
1042 depends on MTRR && NONPROMISC_DEVMEM
1044 Use PAT attributes to setup page level cache control.
1046 PATs are the modern equivalents of MTRRs and are much more
1047 flexible than MTRRs.
1049 Say N here if you see bootup problems (boot crash, boot hang,
1050 spontaneous reboots) or a non-working video driver.
1056 prompt "EFI runtime service support"
1059 This enables the kernel to use EFI runtime services that are
1060 available (such as the EFI variable services).
1062 This option is only useful on systems that have EFI firmware.
1063 In addition, you should use the latest ELILO loader available
1064 at <http://elilo.sourceforge.net> in order to take advantage
1065 of EFI runtime services. However, even with this option, the
1066 resultant kernel should continue to boot on existing non-EFI
1071 prompt "Enable kernel irq balancing"
1072 depends on X86_32 && SMP && X86_IO_APIC
1074 The default yes will allow the kernel to do irq load balancing.
1075 Saying no will keep the kernel from doing irq load balancing.
1079 prompt "Enable seccomp to safely compute untrusted bytecode"
1082 This kernel feature is useful for number crunching applications
1083 that may need to compute untrusted bytecode during their
1084 execution. By using pipes or other transports made available to
1085 the process as file descriptors supporting the read/write
1086 syscalls, it's possible to isolate those applications in
1087 their own address space using seccomp. Once seccomp is
1088 enabled via /proc/<pid>/seccomp, it cannot be disabled
1089 and the task is only allowed to execute a few safe syscalls
1090 defined by each seccomp mode.
1092 If unsure, say Y. Only embedded should say N here.
1094 config CC_STACKPROTECTOR
1095 bool "Enable -fstack-protector buffer overflow detection (EXPERIMENTAL)"
1096 depends on X86_64 && EXPERIMENTAL && BROKEN
1098 This option turns on the -fstack-protector GCC feature. This
1099 feature puts, at the beginning of critical functions, a canary
1100 value on the stack just before the return address, and validates
1101 the value just before actually returning. Stack based buffer
1102 overflows (that need to overwrite this return address) now also
1103 overwrite the canary, which gets detected and the attack is then
1104 neutralized via a kernel panic.
1106 This feature requires gcc version 4.2 or above, or a distribution
1107 gcc with the feature backported. Older versions are automatically
1108 detected and for those versions, this configuration option is ignored.
1110 config CC_STACKPROTECTOR_ALL
1111 bool "Use stack-protector for all functions"
1112 depends on CC_STACKPROTECTOR
1114 Normally, GCC only inserts the canary value protection for
1115 functions that use large-ish on-stack buffers. By enabling
1116 this option, GCC will be asked to do this for ALL functions.
1118 source kernel/Kconfig.hz
1121 bool "kexec system call"
1122 depends on X86_64 || X86_BIOS_REBOOT
1124 kexec is a system call that implements the ability to shutdown your
1125 current kernel, and to start another kernel. It is like a reboot
1126 but it is independent of the system firmware. And like a reboot
1127 you can start any kernel with it, not just Linux.
1129 The name comes from the similarity to the exec system call.
1131 It is an ongoing process to be certain the hardware in a machine
1132 is properly shutdown, so do not be surprised if this code does not
1133 initially work for you. It may help to enable device hotplugging
1134 support. As of this writing the exact hardware interface is
1135 strongly in flux, so no good recommendation can be made.
1138 bool "kernel crash dumps (EXPERIMENTAL)"
1139 depends on EXPERIMENTAL
1140 depends on X86_64 || (X86_32 && HIGHMEM)
1142 Generate crash dump after being started by kexec.
1143 This should be normally only set in special crash dump kernels
1144 which are loaded in the main kernel with kexec-tools into
1145 a specially reserved region and then later executed after
1146 a crash by kdump/kexec. The crash dump kernel must be compiled
1147 to a memory address not used by the main kernel or BIOS using
1148 PHYSICAL_START, or it must be built as a relocatable image
1149 (CONFIG_RELOCATABLE=y).
1150 For more details see Documentation/kdump/kdump.txt
1152 config PHYSICAL_START
1153 hex "Physical address where the kernel is loaded" if (EMBEDDED || CRASH_DUMP)
1154 default "0x1000000" if X86_NUMAQ
1155 default "0x200000" if X86_64
1158 This gives the physical address where the kernel is loaded.
1160 If kernel is a not relocatable (CONFIG_RELOCATABLE=n) then
1161 bzImage will decompress itself to above physical address and
1162 run from there. Otherwise, bzImage will run from the address where
1163 it has been loaded by the boot loader and will ignore above physical
1166 In normal kdump cases one does not have to set/change this option
1167 as now bzImage can be compiled as a completely relocatable image
1168 (CONFIG_RELOCATABLE=y) and be used to load and run from a different
1169 address. This option is mainly useful for the folks who don't want
1170 to use a bzImage for capturing the crash dump and want to use a
1171 vmlinux instead. vmlinux is not relocatable hence a kernel needs
1172 to be specifically compiled to run from a specific memory area
1173 (normally a reserved region) and this option comes handy.
1175 So if you are using bzImage for capturing the crash dump, leave
1176 the value here unchanged to 0x100000 and set CONFIG_RELOCATABLE=y.
1177 Otherwise if you plan to use vmlinux for capturing the crash dump
1178 change this value to start of the reserved region (Typically 16MB
1179 0x1000000). In other words, it can be set based on the "X" value as
1180 specified in the "crashkernel=YM@XM" command line boot parameter
1181 passed to the panic-ed kernel. Typically this parameter is set as
1182 crashkernel=64M@16M. Please take a look at
1183 Documentation/kdump/kdump.txt for more details about crash dumps.
1185 Usage of bzImage for capturing the crash dump is recommended as
1186 one does not have to build two kernels. Same kernel can be used
1187 as production kernel and capture kernel. Above option should have
1188 gone away after relocatable bzImage support is introduced. But it
1189 is present because there are users out there who continue to use
1190 vmlinux for dump capture. This option should go away down the
1193 Don't change this unless you know what you are doing.
1196 bool "Build a relocatable kernel (EXPERIMENTAL)"
1197 depends on EXPERIMENTAL
1199 This builds a kernel image that retains relocation information
1200 so it can be loaded someplace besides the default 1MB.
1201 The relocations tend to make the kernel binary about 10% larger,
1202 but are discarded at runtime.
1204 One use is for the kexec on panic case where the recovery kernel
1205 must live at a different physical address than the primary
1208 Note: If CONFIG_RELOCATABLE=y, then the kernel runs from the address
1209 it has been loaded at and the compile time physical address
1210 (CONFIG_PHYSICAL_START) is ignored.
1212 config PHYSICAL_ALIGN
1214 prompt "Alignment value to which kernel should be aligned" if X86_32
1215 default "0x100000" if X86_32
1216 default "0x200000" if X86_64
1217 range 0x2000 0x400000
1219 This value puts the alignment restrictions on physical address
1220 where kernel is loaded and run from. Kernel is compiled for an
1221 address which meets above alignment restriction.
1223 If bootloader loads the kernel at a non-aligned address and
1224 CONFIG_RELOCATABLE is set, kernel will move itself to nearest
1225 address aligned to above value and run from there.
1227 If bootloader loads the kernel at a non-aligned address and
1228 CONFIG_RELOCATABLE is not set, kernel will ignore the run time
1229 load address and decompress itself to the address it has been
1230 compiled for and run from there. The address for which kernel is
1231 compiled already meets above alignment restrictions. Hence the
1232 end result is that kernel runs from a physical address meeting
1233 above alignment restrictions.
1235 Don't change this unless you know what you are doing.
1238 bool "Support for suspend on SMP and hot-pluggable CPUs (EXPERIMENTAL)"
1239 depends on SMP && HOTPLUG && EXPERIMENTAL && !X86_VOYAGER
1241 Say Y here to experiment with turning CPUs off and on, and to
1242 enable suspend on SMP systems. CPUs can be controlled through
1243 /sys/devices/system/cpu.
1244 Say N if you want to disable CPU hotplug and don't need to
1249 prompt "Compat VDSO support"
1250 depends on X86_32 || IA32_EMULATION
1252 Map the 32-bit VDSO to the predictable old-style address too.
1254 Say N here if you are running a sufficiently recent glibc
1255 version (2.3.3 or later), to remove the high-mapped
1256 VDSO mapping and to exclusively use the randomized VDSO.
1262 config ARCH_ENABLE_MEMORY_HOTPLUG
1264 depends on X86_64 || (X86_32 && HIGHMEM)
1266 config HAVE_ARCH_EARLY_PFN_TO_NID
1270 menu "Power management options"
1271 depends on !X86_VOYAGER
1273 config ARCH_HIBERNATION_HEADER
1275 depends on X86_64 && HIBERNATION
1277 source "kernel/power/Kconfig"
1279 source "drivers/acpi/Kconfig"
1284 depends on APM || APM_MODULE
1287 tristate "APM (Advanced Power Management) BIOS support"
1288 depends on X86_32 && PM_SLEEP && !X86_VISWS
1290 APM is a BIOS specification for saving power using several different
1291 techniques. This is mostly useful for battery powered laptops with
1292 APM compliant BIOSes. If you say Y here, the system time will be
1293 reset after a RESUME operation, the /proc/apm device will provide
1294 battery status information, and user-space programs will receive
1295 notification of APM "events" (e.g. battery status change).
1297 If you select "Y" here, you can disable actual use of the APM
1298 BIOS by passing the "apm=off" option to the kernel at boot time.
1300 Note that the APM support is almost completely disabled for
1301 machines with more than one CPU.
1303 In order to use APM, you will need supporting software. For location
1304 and more information, read <file:Documentation/power/pm.txt> and the
1305 Battery Powered Linux mini-HOWTO, available from
1306 <http://www.tldp.org/docs.html#howto>.
1308 This driver does not spin down disk drives (see the hdparm(8)
1309 manpage ("man 8 hdparm") for that), and it doesn't turn off
1310 VESA-compliant "green" monitors.
1312 This driver does not support the TI 4000M TravelMate and the ACER
1313 486/DX4/75 because they don't have compliant BIOSes. Many "green"
1314 desktop machines also don't have compliant BIOSes, and this driver
1315 may cause those machines to panic during the boot phase.
1317 Generally, if you don't have a battery in your machine, there isn't
1318 much point in using this driver and you should say N. If you get
1319 random kernel OOPSes or reboots that don't seem to be related to
1320 anything, try disabling/enabling this option (or disabling/enabling
1323 Some other things you should try when experiencing seemingly random,
1326 1) make sure that you have enough swap space and that it is
1328 2) pass the "no-hlt" option to the kernel
1329 3) switch on floating point emulation in the kernel and pass
1330 the "no387" option to the kernel
1331 4) pass the "floppy=nodma" option to the kernel
1332 5) pass the "mem=4M" option to the kernel (thereby disabling
1333 all but the first 4 MB of RAM)
1334 6) make sure that the CPU is not over clocked.
1335 7) read the sig11 FAQ at <http://www.bitwizard.nl/sig11/>
1336 8) disable the cache from your BIOS settings
1337 9) install a fan for the video card or exchange video RAM
1338 10) install a better fan for the CPU
1339 11) exchange RAM chips
1340 12) exchange the motherboard.
1342 To compile this driver as a module, choose M here: the
1343 module will be called apm.
1347 config APM_IGNORE_USER_SUSPEND
1348 bool "Ignore USER SUSPEND"
1350 This option will ignore USER SUSPEND requests. On machines with a
1351 compliant APM BIOS, you want to say N. However, on the NEC Versa M
1352 series notebooks, it is necessary to say Y because of a BIOS bug.
1354 config APM_DO_ENABLE
1355 bool "Enable PM at boot time"
1357 Enable APM features at boot time. From page 36 of the APM BIOS
1358 specification: "When disabled, the APM BIOS does not automatically
1359 power manage devices, enter the Standby State, enter the Suspend
1360 State, or take power saving steps in response to CPU Idle calls."
1361 This driver will make CPU Idle calls when Linux is idle (unless this
1362 feature is turned off -- see "Do CPU IDLE calls", below). This
1363 should always save battery power, but more complicated APM features
1364 will be dependent on your BIOS implementation. You may need to turn
1365 this option off if your computer hangs at boot time when using APM
1366 support, or if it beeps continuously instead of suspending. Turn
1367 this off if you have a NEC UltraLite Versa 33/C or a Toshiba
1368 T400CDT. This is off by default since most machines do fine without
1372 bool "Make CPU Idle calls when idle"
1374 Enable calls to APM CPU Idle/CPU Busy inside the kernel's idle loop.
1375 On some machines, this can activate improved power savings, such as
1376 a slowed CPU clock rate, when the machine is idle. These idle calls
1377 are made after the idle loop has run for some length of time (e.g.,
1378 333 mS). On some machines, this will cause a hang at boot time or
1379 whenever the CPU becomes idle. (On machines with more than one CPU,
1380 this option does nothing.)
1382 config APM_DISPLAY_BLANK
1383 bool "Enable console blanking using APM"
1385 Enable console blanking using the APM. Some laptops can use this to
1386 turn off the LCD backlight when the screen blanker of the Linux
1387 virtual console blanks the screen. Note that this is only used by
1388 the virtual console screen blanker, and won't turn off the backlight
1389 when using the X Window system. This also doesn't have anything to
1390 do with your VESA-compliant power-saving monitor. Further, this
1391 option doesn't work for all laptops -- it might not turn off your
1392 backlight at all, or it might print a lot of errors to the console,
1393 especially if you are using gpm.
1395 config APM_ALLOW_INTS
1396 bool "Allow interrupts during APM BIOS calls"
1398 Normally we disable external interrupts while we are making calls to
1399 the APM BIOS as a measure to lessen the effects of a badly behaving
1400 BIOS implementation. The BIOS should reenable interrupts if it
1401 needs to. Unfortunately, some BIOSes do not -- especially those in
1402 many of the newer IBM Thinkpads. If you experience hangs when you
1403 suspend, try setting this to Y. Otherwise, say N.
1405 config APM_REAL_MODE_POWER_OFF
1406 bool "Use real mode APM BIOS call to power off"
1408 Use real mode APM BIOS calls to switch off the computer. This is
1409 a work-around for a number of buggy BIOSes. Switch this option on if
1410 your computer crashes instead of powering off properly.
1414 source "arch/x86/kernel/cpu/cpufreq/Kconfig"
1416 source "drivers/cpuidle/Kconfig"
1421 menu "Bus options (PCI etc.)"
1424 bool "PCI support" if !X86_VISWS && !X86_VSMP
1425 depends on !X86_VOYAGER
1427 select ARCH_SUPPORTS_MSI if (X86_LOCAL_APIC && X86_IO_APIC)
1429 Find out whether you have a PCI motherboard. PCI is the name of a
1430 bus system, i.e. the way the CPU talks to the other stuff inside
1431 your box. Other bus systems are ISA, EISA, MicroChannel (MCA) or
1432 VESA. If you have PCI, say Y, otherwise N.
1435 prompt "PCI access mode"
1436 depends on X86_32 && PCI && !X86_VISWS
1439 On PCI systems, the BIOS can be used to detect the PCI devices and
1440 determine their configuration. However, some old PCI motherboards
1441 have BIOS bugs and may crash if this is done. Also, some embedded
1442 PCI-based systems don't have any BIOS at all. Linux can also try to
1443 detect the PCI hardware directly without using the BIOS.
1445 With this option, you can specify how Linux should detect the
1446 PCI devices. If you choose "BIOS", the BIOS will be used,
1447 if you choose "Direct", the BIOS won't be used, and if you
1448 choose "MMConfig", then PCI Express MMCONFIG will be used.
1449 If you choose "Any", the kernel will try MMCONFIG, then the
1450 direct access method and falls back to the BIOS if that doesn't
1451 work. If unsure, go with the default, which is "Any".
1456 config PCI_GOMMCONFIG
1469 depends on X86_32 && !X86_VISWS && PCI && (PCI_GOBIOS || PCI_GOANY)
1471 # x86-64 doesn't support PCI BIOS access from long mode so always go direct.
1474 depends on PCI && (X86_64 || (PCI_GODIRECT || PCI_GOANY) || X86_VISWS)
1478 depends on X86_32 && PCI && ACPI && (PCI_GOMMCONFIG || PCI_GOANY)
1485 bool "Support mmconfig PCI config space access"
1486 depends on X86_64 && PCI && ACPI
1489 bool "Support for DMA Remapping Devices (EXPERIMENTAL)"
1490 depends on X86_64 && PCI_MSI && ACPI && EXPERIMENTAL
1492 DMA remapping (DMAR) devices support enables independent address
1493 translations for Direct Memory Access (DMA) from devices.
1494 These DMA remapping devices are reported via ACPI tables
1495 and include PCI device scope covered by these DMA
1500 prompt "Support for Graphics workaround"
1503 Current Graphics drivers tend to use physical address
1504 for DMA and avoid using DMA APIs. Setting this config
1505 option permits the IOMMU driver to set a unity map for
1506 all the OS-visible memory. Hence the driver can continue
1507 to use physical addresses for DMA.
1509 config DMAR_FLOPPY_WA
1513 Floppy disk drivers are know to bypass DMA API calls
1514 thereby failing to work when IOMMU is enabled. This
1515 workaround will setup a 1:1 mapping for the first
1516 16M to make floppy (an ISA device) work.
1518 source "drivers/pci/pcie/Kconfig"
1520 source "drivers/pci/Kconfig"
1522 # x86_64 have no ISA slots, but do have ISA-style DMA.
1530 depends on !(X86_VOYAGER || X86_VISWS)
1532 Find out whether you have ISA slots on your motherboard. ISA is the
1533 name of a bus system, i.e. the way the CPU talks to the other stuff
1534 inside your box. Other bus systems are PCI, EISA, MicroChannel
1535 (MCA) or VESA. ISA is an older system, now being displaced by PCI;
1536 newer boards don't support it. If you have ISA, say Y, otherwise N.
1542 The Extended Industry Standard Architecture (EISA) bus was
1543 developed as an open alternative to the IBM MicroChannel bus.
1545 The EISA bus provided some of the features of the IBM MicroChannel
1546 bus while maintaining backward compatibility with cards made for
1547 the older ISA bus. The EISA bus saw limited use between 1988 and
1548 1995 when it was made obsolete by the PCI bus.
1550 Say Y here if you are building a kernel for an EISA-based machine.
1554 source "drivers/eisa/Kconfig"
1557 bool "MCA support" if !(X86_VISWS || X86_VOYAGER)
1558 default y if X86_VOYAGER
1560 MicroChannel Architecture is found in some IBM PS/2 machines and
1561 laptops. It is a bus system similar to PCI or ISA. See
1562 <file:Documentation/mca.txt> (and especially the web page given
1563 there) before attempting to build an MCA bus kernel.
1565 source "drivers/mca/Kconfig"
1568 tristate "NatSemi SCx200 support"
1569 depends on !X86_VOYAGER
1571 This provides basic support for National Semiconductor's
1572 (now AMD's) Geode processors. The driver probes for the
1573 PCI-IDs of several on-chip devices, so its a good dependency
1574 for other scx200_* drivers.
1576 If compiled as a module, the driver is named scx200.
1578 config SCx200HR_TIMER
1579 tristate "NatSemi SCx200 27MHz High-Resolution Timer Support"
1580 depends on SCx200 && GENERIC_TIME
1583 This driver provides a clocksource built upon the on-chip
1584 27MHz high-resolution timer. Its also a workaround for
1585 NSC Geode SC-1100's buggy TSC, which loses time when the
1586 processor goes idle (as is done by the scheduler). The
1587 other workaround is idle=poll boot option.
1589 config GEODE_MFGPT_TIMER
1591 prompt "Geode Multi-Function General Purpose Timer (MFGPT) events"
1592 depends on MGEODE_LX && GENERIC_TIME && GENERIC_CLOCKEVENTS
1594 This driver provides a clock event source based on the MFGPT
1595 timer(s) in the CS5535 and CS5536 companion chip for the geode.
1596 MFGPTs have a better resolution and max interval than the
1597 generic PIT, and are suitable for use as high-res timers.
1603 depends on AGP_AMD64 || (X86_64 && (GART_IOMMU || (PCI && NUMA)))
1605 source "drivers/pcmcia/Kconfig"
1607 source "drivers/pci/hotplug/Kconfig"
1612 menu "Executable file formats / Emulations"
1614 source "fs/Kconfig.binfmt"
1616 config IA32_EMULATION
1617 bool "IA32 Emulation"
1619 select COMPAT_BINFMT_ELF
1621 Include code to run 32-bit programs under a 64-bit kernel. You should
1622 likely turn this on, unless you're 100% sure that you don't have any
1623 32-bit programs left.
1626 tristate "IA32 a.out support"
1627 depends on IA32_EMULATION && ARCH_SUPPORTS_AOUT
1629 Support old a.out binaries in the 32bit emulation.
1633 depends on IA32_EMULATION
1635 config COMPAT_FOR_U64_ALIGNMENT
1639 config SYSVIPC_COMPAT
1641 depends on X86_64 && COMPAT && SYSVIPC
1646 source "net/Kconfig"
1648 source "drivers/Kconfig"
1650 source "drivers/firmware/Kconfig"
1654 source "arch/x86/Kconfig.debug"
1656 source "security/Kconfig"
1658 source "crypto/Kconfig"
1660 source "arch/x86/kvm/Kconfig"
1662 source "lib/Kconfig"
projects / deliverable / linux.git / blob