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)
28 config GENERIC_LOCKBREAK
34 config GENERIC_CMOS_UPDATE
37 config CLOCKSOURCE_WATCHDOG
40 config GENERIC_CLOCKEVENTS
43 config GENERIC_CLOCKEVENTS_BROADCAST
45 depends on X86_64 || (X86_32 && X86_LOCAL_APIC)
47 config LOCKDEP_SUPPORT
50 config STACKTRACE_SUPPORT
53 config HAVE_LATENCYTOP_SUPPORT
56 config FAST_CMPXCHG_LOCAL
69 config GENERIC_ISA_DMA
79 config GENERIC_HWEIGHT
85 config ARCH_MAY_HAVE_PC_FDC
91 config RWSEM_GENERIC_SPINLOCK
94 config RWSEM_XCHGADD_ALGORITHM
97 config ARCH_HAS_ILOG2_U32
100 config ARCH_HAS_ILOG2_U64
103 config ARCH_HAS_CPU_IDLE_WAIT
106 config GENERIC_CALIBRATE_DELAY
109 config GENERIC_TIME_VSYSCALL
113 config ARCH_HAS_CPU_RELAX
116 config HAVE_SETUP_PER_CPU_AREA
117 def_bool X86_64 || (X86_SMP && !X86_VOYAGER)
119 config ARCH_HIBERNATION_POSSIBLE
121 depends on !SMP || !X86_VOYAGER
123 config ARCH_SUSPEND_POSSIBLE
125 depends on !X86_VOYAGER
131 config ARCH_POPULATES_NODE_MAP
138 config ARCH_SUPPORTS_AOUT
141 # Use the generic interrupt handling code in kernel/irq/:
142 config GENERIC_HARDIRQS
146 config GENERIC_IRQ_PROBE
150 config GENERIC_PENDING_IRQ
152 depends on GENERIC_HARDIRQS && SMP
157 depends on SMP && ((X86_32 && !X86_VOYAGER) || X86_64)
162 depends on X86_32 && SMP
166 depends on X86_64 && SMP
171 depends on (X86_32 && !(X86_VISWS || X86_VOYAGER)) || X86_64
174 config X86_BIOS_REBOOT
176 depends on X86_32 && !(X86_VISWS || X86_VOYAGER)
179 config X86_TRAMPOLINE
181 depends on X86_SMP || (X86_VOYAGER && SMP) || (64BIT && ACPI_SLEEP)
186 source "init/Kconfig"
188 menu "Processor type and features"
190 source "kernel/time/Kconfig"
193 bool "Symmetric multi-processing support"
195 This enables support for systems with more than one CPU. If you have
196 a system with only one CPU, like most personal computers, say N. If
197 you have a system with more than one CPU, say Y.
199 If you say N here, the kernel will run on single and multiprocessor
200 machines, but will use only one CPU of a multiprocessor machine. If
201 you say Y here, the kernel will run on many, but not all,
202 singleprocessor machines. On a singleprocessor machine, the kernel
203 will run faster if you say N here.
205 Note that if you say Y here and choose architecture "586" or
206 "Pentium" under "Processor family", the kernel will not work on 486
207 architectures. Similarly, multiprocessor kernels for the "PPro"
208 architecture may not work on all Pentium based boards.
210 People using multiprocessor machines who say Y here should also say
211 Y to "Enhanced Real Time Clock Support", below. The "Advanced Power
212 Management" code will be disabled if you say Y here.
214 See also <file:Documentation/i386/IO-APIC.txt>,
215 <file:Documentation/nmi_watchdog.txt> and the SMP-HOWTO available at
216 <http://www.tldp.org/docs.html#howto>.
218 If you don't know what to do here, say N.
221 prompt "Subarchitecture Type"
227 Choose this option if your computer is a standard PC or compatible.
233 Select this for an AMD Elan processor.
235 Do not use this option for K6/Athlon/Opteron processors!
237 If unsure, choose "PC-compatible" instead.
241 depends on X86_32 && (SMP || BROKEN)
243 Voyager is an MCA-based 32-way capable SMP architecture proprietary
244 to NCR Corp. Machine classes 345x/35xx/4100/51xx are Voyager-based.
248 If you do not specifically know you have a Voyager based machine,
249 say N here, otherwise the kernel you build will not be bootable.
252 bool "NUMAQ (IBM/Sequent)"
253 depends on SMP && X86_32
256 This option is used for getting Linux to run on a (IBM/Sequent) NUMA
257 multiquad box. This changes the way that processors are bootstrapped,
258 and uses Clustered Logical APIC addressing mode instead of Flat Logical.
259 You will need a new lynxer.elf file to flash your firmware with - send
260 email to <Martin.Bligh@us.ibm.com>.
263 bool "Summit/EXA (IBM x440)"
264 depends on X86_32 && SMP
266 This option is needed for IBM systems that use the Summit/EXA chipset.
267 In particular, it is needed for the x440.
269 If you don't have one of these computers, you should say N here.
270 If you want to build a NUMA kernel, you must select ACPI.
273 bool "Support for other sub-arch SMP systems with more than 8 CPUs"
274 depends on X86_32 && SMP
276 This option is needed for the systems that have more than 8 CPUs
277 and if the system is not of any sub-arch type above.
279 If you don't have such a system, you should say N here.
282 bool "SGI 320/540 (Visual Workstation)"
285 The SGI Visual Workstation series is an IA32-based workstation
286 based on SGI systems chips with some legacy PC hardware attached.
288 Say Y here to create a kernel to run on the SGI 320 or 540.
290 A kernel compiled for the Visual Workstation will not run on PCs
291 and vice versa. See <file:Documentation/sgi-visws.txt> for details.
293 config X86_GENERICARCH
294 bool "Generic architecture (Summit, bigsmp, ES7000, default)"
297 This option compiles in the Summit, bigsmp, ES7000, default subarchitectures.
298 It is intended for a generic binary kernel.
299 If you want a NUMA kernel, select ACPI. We need SRAT for NUMA.
302 bool "Support for Unisys ES7000 IA32 series"
303 depends on X86_32 && SMP
305 Support for Unisys ES7000 systems. Say 'Y' here if this kernel is
306 supposed to run on an IA32-based Unisys ES7000 system.
307 Only choose this option if you have such a system, otherwise you
311 bool "RDC R-321x SoC"
314 select X86_REBOOTFIXUPS
319 This option is needed for RDC R-321x system-on-chip, also known
321 If you don't have one of these chips, you should say N here.
324 bool "Support for ScaleMP vSMP"
328 Support for ScaleMP vSMP systems. Say 'Y' here if this kernel is
329 supposed to run on these EM64T-based machines. Only choose this option
330 if you have one of these machines.
334 config SCHED_NO_NO_OMIT_FRAME_POINTER
336 prompt "Single-depth WCHAN output"
339 Calculate simpler /proc/<PID>/wchan values. If this option
340 is disabled then wchan values will recurse back to the
341 caller function. This provides more accurate wchan values,
342 at the expense of slightly more scheduling overhead.
344 If in doubt, say "Y".
346 menuconfig PARAVIRT_GUEST
347 bool "Paravirtualized guest support"
349 Say Y here to get to see options related to running Linux under
350 various hypervisors. This option alone does not add any kernel code.
352 If you say N, all options in this submenu will be skipped and disabled.
356 source "arch/x86/xen/Kconfig"
359 bool "VMI Guest support"
362 depends on !(X86_VISWS || X86_VOYAGER)
364 VMI provides a paravirtualized interface to the VMware ESX server
365 (it could be used by other hypervisors in theory too, but is not
366 at the moment), by linking the kernel to a GPL-ed ROM module
367 provided by the hypervisor.
369 source "arch/x86/lguest/Kconfig"
372 bool "Enable paravirtualization code"
373 depends on !(X86_VISWS || X86_VOYAGER)
375 This changes the kernel so it can modify itself when it is run
376 under a hypervisor, potentially improving performance significantly
377 over full virtualization. However, when run without a hypervisor
378 the kernel is theoretically slower and slightly larger.
382 config MEMTEST_BOOTPARAM
383 bool "Memtest boot parameter"
387 This option adds a kernel parameter 'memtest', which allows memtest
388 to be disabled at boot. If this option is selected, memtest
389 functionality can be disabled with memtest=0 on the kernel
390 command line. The purpose of this option is to allow a single
391 kernel image to be distributed with memtest built in, but not
394 If you are unsure how to answer this question, answer Y.
396 config MEMTEST_BOOTPARAM_VALUE
397 int "Memtest boot parameter default value (0-4)"
398 depends on MEMTEST_BOOTPARAM
402 This option sets the default value for the kernel parameter
403 'memtest', which allows memtest to be disabled at boot. If this
404 option is set to 0 (zero), the memtest kernel parameter will
405 default to 0, disabling memtest at bootup. If this option is
406 set to 4, the memtest kernel parameter will default to 4,
407 enabling memtest at bootup, and use that as pattern number.
409 If you are unsure how to answer this question, answer 0.
413 depends on X86_32 && ACPI && NUMA && (X86_SUMMIT || X86_GENERICARCH)
416 config HAVE_ARCH_PARSE_SRAT
420 config X86_SUMMIT_NUMA
422 depends on X86_32 && NUMA && (X86_SUMMIT || X86_GENERICARCH)
424 config X86_CYCLONE_TIMER
426 depends on X86_32 && X86_SUMMIT || X86_GENERICARCH
428 config ES7000_CLUSTERED_APIC
430 depends on SMP && X86_ES7000 && MPENTIUMIII
432 source "arch/x86/Kconfig.cpu"
436 prompt "HPET Timer Support" if X86_32
438 Use the IA-PC HPET (High Precision Event Timer) to manage
439 time in preference to the PIT and RTC, if a HPET is
441 HPET is the next generation timer replacing legacy 8254s.
442 The HPET provides a stable time base on SMP
443 systems, unlike the TSC, but it is more expensive to access,
444 as it is off-chip. You can find the HPET spec at
445 <http://www.intel.com/hardwaredesign/hpetspec.htm>.
447 You can safely choose Y here. However, HPET will only be
448 activated if the platform and the BIOS support this feature.
449 Otherwise the 8254 will be used for timing services.
451 Choose N to continue using the legacy 8254 timer.
453 config HPET_EMULATE_RTC
455 depends on HPET_TIMER && (RTC=y || RTC=m || RTC_DRV_CMOS=m || RTC_DRV_CMOS=y)
457 # Mark as embedded because too many people got it wrong.
458 # The code disables itself when not needed.
460 bool "GART IOMMU support" if EMBEDDED
464 depends on X86_64 && PCI
466 Support for full DMA access of devices with 32bit memory access only
467 on systems with more than 3GB. This is usually needed for USB,
468 sound, many IDE/SATA chipsets and some other devices.
469 Provides a driver for the AMD Athlon64/Opteron/Turion/Sempron GART
470 based hardware IOMMU and a software bounce buffer based IOMMU used
471 on Intel systems and as fallback.
472 The code is only active when needed (enough memory and limited
473 device) unless CONFIG_IOMMU_DEBUG or iommu=force is specified
477 bool "IBM Calgary IOMMU support"
479 depends on X86_64 && PCI && EXPERIMENTAL
481 Support for hardware IOMMUs in IBM's xSeries x366 and x460
482 systems. Needed to run systems with more than 3GB of memory
483 properly with 32-bit PCI devices that do not support DAC
484 (Double Address Cycle). Calgary also supports bus level
485 isolation, where all DMAs pass through the IOMMU. This
486 prevents them from going anywhere except their intended
487 destination. This catches hard-to-find kernel bugs and
488 mis-behaving drivers and devices that do not use the DMA-API
489 properly to set up their DMA buffers. The IOMMU can be
490 turned off at boot time with the iommu=off parameter.
491 Normally the kernel will make the right choice by itself.
494 config CALGARY_IOMMU_ENABLED_BY_DEFAULT
496 prompt "Should Calgary be enabled by default?"
497 depends on CALGARY_IOMMU
499 Should Calgary be enabled by default? if you choose 'y', Calgary
500 will be used (if it exists). If you choose 'n', Calgary will not be
501 used even if it exists. If you choose 'n' and would like to use
502 Calgary anyway, pass 'iommu=calgary' on the kernel command line.
506 def_bool (CALGARY_IOMMU || GART_IOMMU)
508 # need this always selected by IOMMU for the VIA workaround
512 Support for software bounce buffers used on x86-64 systems
513 which don't have a hardware IOMMU (e.g. the current generation
514 of Intel's x86-64 CPUs). Using this PCI devices which can only
515 access 32-bits of memory can be used on systems with more than
516 3 GB of memory. If unsure, say Y.
520 int "Maximum number of CPUs (2-255)"
523 default "32" if X86_NUMAQ || X86_SUMMIT || X86_BIGSMP || X86_ES7000
526 This allows you to specify the maximum number of CPUs which this
527 kernel will support. The maximum supported value is 255 and the
528 minimum value which makes sense is 2.
530 This is purely to save memory - each supported CPU adds
531 approximately eight kilobytes to the kernel image.
534 bool "SMT (Hyperthreading) scheduler support"
537 SMT scheduler support improves the CPU scheduler's decision making
538 when dealing with Intel Pentium 4 chips with HyperThreading at a
539 cost of slightly increased overhead in some places. If unsure say
544 prompt "Multi-core scheduler support"
547 Multi-core scheduler support improves the CPU scheduler's decision
548 making when dealing with multi-core CPU chips at a cost of slightly
549 increased overhead in some places. If unsure say N here.
551 source "kernel/Kconfig.preempt"
554 bool "Local APIC support on uniprocessors"
555 depends on X86_32 && !SMP && !(X86_VISWS || X86_VOYAGER || X86_GENERICARCH)
557 A local APIC (Advanced Programmable Interrupt Controller) is an
558 integrated interrupt controller in the CPU. If you have a single-CPU
559 system which has a processor with a local APIC, you can say Y here to
560 enable and use it. If you say Y here even though your machine doesn't
561 have a local APIC, then the kernel will still run with no slowdown at
562 all. The local APIC supports CPU-generated self-interrupts (timer,
563 performance counters), and the NMI watchdog which detects hard
567 bool "IO-APIC support on uniprocessors"
568 depends on X86_UP_APIC
570 An IO-APIC (I/O Advanced Programmable Interrupt Controller) is an
571 SMP-capable replacement for PC-style interrupt controllers. Most
572 SMP systems and many recent uniprocessor systems have one.
574 If you have a single-CPU system with an IO-APIC, you can say Y here
575 to use it. If you say Y here even though your machine doesn't have
576 an IO-APIC, then the kernel will still run with no slowdown at all.
578 config X86_LOCAL_APIC
580 depends on X86_64 || (X86_32 && (X86_UP_APIC || ((X86_VISWS || SMP) && !X86_VOYAGER) || X86_GENERICARCH))
584 depends on X86_64 || (X86_32 && (X86_UP_IOAPIC || (SMP && !(X86_VISWS || X86_VOYAGER)) || X86_GENERICARCH))
586 config X86_VISWS_APIC
588 depends on X86_32 && X86_VISWS
591 bool "Machine Check Exception"
592 depends on !X86_VOYAGER
594 Machine Check Exception support allows the processor to notify the
595 kernel if it detects a problem (e.g. overheating, component failure).
596 The action the kernel takes depends on the severity of the problem,
597 ranging from a warning message on the console, to halting the machine.
598 Your processor must be a Pentium or newer to support this - check the
599 flags in /proc/cpuinfo for mce. Note that some older Pentium systems
600 have a design flaw which leads to false MCE events - hence MCE is
601 disabled on all P5 processors, unless explicitly enabled with "mce"
602 as a boot argument. Similarly, if MCE is built in and creates a
603 problem on some new non-standard machine, you can boot with "nomce"
604 to disable it. MCE support simply ignores non-MCE processors like
605 the 386 and 486, so nearly everyone can say Y here.
609 prompt "Intel MCE features"
610 depends on X86_64 && X86_MCE && X86_LOCAL_APIC
612 Additional support for intel specific MCE features such as
617 prompt "AMD MCE features"
618 depends on X86_64 && X86_MCE && X86_LOCAL_APIC
620 Additional support for AMD specific MCE features such as
621 the DRAM Error Threshold.
623 config X86_MCE_NONFATAL
624 tristate "Check for non-fatal errors on AMD Athlon/Duron / Intel Pentium 4"
625 depends on X86_32 && X86_MCE
627 Enabling this feature starts a timer that triggers every 5 seconds which
628 will look at the machine check registers to see if anything happened.
629 Non-fatal problems automatically get corrected (but still logged).
630 Disable this if you don't want to see these messages.
631 Seeing the messages this option prints out may be indicative of dying
632 or out-of-spec (ie, overclocked) hardware.
633 This option only does something on certain CPUs.
634 (AMD Athlon/Duron and Intel Pentium 4)
636 config X86_MCE_P4THERMAL
637 bool "check for P4 thermal throttling interrupt."
638 depends on X86_32 && X86_MCE && (X86_UP_APIC || SMP) && !X86_VISWS
640 Enabling this feature will cause a message to be printed when the P4
641 enters thermal throttling.
644 bool "Enable VM86 support" if EMBEDDED
648 This option is required by programs like DOSEMU to run 16-bit legacy
649 code on X86 processors. It also may be needed by software like
650 XFree86 to initialize some video cards via BIOS. Disabling this
651 option saves about 6k.
654 tristate "Toshiba Laptop support"
657 This adds a driver to safely access the System Management Mode of
658 the CPU on Toshiba portables with a genuine Toshiba BIOS. It does
659 not work on models with a Phoenix BIOS. The System Management Mode
660 is used to set the BIOS and power saving options on Toshiba portables.
662 For information on utilities to make use of this driver see the
663 Toshiba Linux utilities web site at:
664 <http://www.buzzard.org.uk/toshiba/>.
666 Say Y if you intend to run this kernel on a Toshiba portable.
670 tristate "Dell laptop support"
672 This adds a driver to safely access the System Management Mode
673 of the CPU on the Dell Inspiron 8000. The System Management Mode
674 is used to read cpu temperature and cooling fan status and to
675 control the fans on the I8K portables.
677 This driver has been tested only on the Inspiron 8000 but it may
678 also work with other Dell laptops. You can force loading on other
679 models by passing the parameter `force=1' to the module. Use at
682 For information on utilities to make use of this driver see the
683 I8K Linux utilities web site at:
684 <http://people.debian.org/~dz/i8k/>
686 Say Y if you intend to run this kernel on a Dell Inspiron 8000.
689 config X86_REBOOTFIXUPS
691 prompt "Enable X86 board specific fixups for reboot"
692 depends on X86_32 && X86
694 This enables chipset and/or board specific fixups to be done
695 in order to get reboot to work correctly. This is only needed on
696 some combinations of hardware and BIOS. The symptom, for which
697 this config is intended, is when reboot ends with a stalled/hung
700 Currently, the only fixup is for the Geode machines using
701 CS5530A and CS5536 chipsets and the RDC R-321x SoC.
703 Say Y if you want to enable the fixup. Currently, it's safe to
704 enable this option even if you don't need it.
708 tristate "/dev/cpu/microcode - Intel IA32 CPU microcode support"
711 If you say Y here, you will be able to update the microcode on
712 Intel processors in the IA32 family, e.g. Pentium Pro, Pentium II,
713 Pentium III, Pentium 4, Xeon etc. You will obviously need the
714 actual microcode binary data itself which is not shipped with the
717 For latest news and information on obtaining all the required
718 ingredients for this driver, check:
719 <http://www.urbanmyth.org/microcode/>.
721 To compile this driver as a module, choose M here: the
722 module will be called microcode.
724 config MICROCODE_OLD_INTERFACE
729 tristate "/dev/cpu/*/msr - Model-specific register support"
731 This device gives privileged processes access to the x86
732 Model-Specific Registers (MSRs). It is a character device with
733 major 202 and minors 0 to 31 for /dev/cpu/0/msr to /dev/cpu/31/msr.
734 MSR accesses are directed to a specific CPU on multi-processor
738 tristate "/dev/cpu/*/cpuid - CPU information support"
740 This device gives processes access to the x86 CPUID instruction to
741 be executed on a specific processor. It is a character device
742 with major 203 and minors 0 to 31 for /dev/cpu/0/cpuid to
746 prompt "High Memory Support"
747 default HIGHMEM4G if !X86_NUMAQ
748 default HIGHMEM64G if X86_NUMAQ
753 depends on !X86_NUMAQ
755 Linux can use up to 64 Gigabytes of physical memory on x86 systems.
756 However, the address space of 32-bit x86 processors is only 4
757 Gigabytes large. That means that, if you have a large amount of
758 physical memory, not all of it can be "permanently mapped" by the
759 kernel. The physical memory that's not permanently mapped is called
762 If you are compiling a kernel which will never run on a machine with
763 more than 1 Gigabyte total physical RAM, answer "off" here (default
764 choice and suitable for most users). This will result in a "3GB/1GB"
765 split: 3GB are mapped so that each process sees a 3GB virtual memory
766 space and the remaining part of the 4GB virtual memory space is used
767 by the kernel to permanently map as much physical memory as
770 If the machine has between 1 and 4 Gigabytes physical RAM, then
773 If more than 4 Gigabytes is used then answer "64GB" here. This
774 selection turns Intel PAE (Physical Address Extension) mode on.
775 PAE implements 3-level paging on IA32 processors. PAE is fully
776 supported by Linux, PAE mode is implemented on all recent Intel
777 processors (Pentium Pro and better). NOTE: If you say "64GB" here,
778 then the kernel will not boot on CPUs that don't support PAE!
780 The actual amount of total physical memory will either be
781 auto detected or can be forced by using a kernel command line option
782 such as "mem=256M". (Try "man bootparam" or see the documentation of
783 your boot loader (lilo or loadlin) about how to pass options to the
784 kernel at boot time.)
786 If unsure, say "off".
790 depends on !X86_NUMAQ
792 Select this if you have a 32-bit processor and between 1 and 4
793 gigabytes of physical RAM.
797 depends on !M386 && !M486
800 Select this if you have a 32-bit processor and more than 4
801 gigabytes of physical RAM.
806 depends on EXPERIMENTAL
807 prompt "Memory split" if EMBEDDED
811 Select the desired split between kernel and user memory.
813 If the address range available to the kernel is less than the
814 physical memory installed, the remaining memory will be available
815 as "high memory". Accessing high memory is a little more costly
816 than low memory, as it needs to be mapped into the kernel first.
817 Note that increasing the kernel address space limits the range
818 available to user programs, making the address space there
819 tighter. Selecting anything other than the default 3G/1G split
820 will also likely make your kernel incompatible with binary-only
823 If you are not absolutely sure what you are doing, leave this
827 bool "3G/1G user/kernel split"
828 config VMSPLIT_3G_OPT
830 bool "3G/1G user/kernel split (for full 1G low memory)"
832 bool "2G/2G user/kernel split"
833 config VMSPLIT_2G_OPT
835 bool "2G/2G user/kernel split (for full 2G low memory)"
837 bool "1G/3G user/kernel split"
842 default 0xB0000000 if VMSPLIT_3G_OPT
843 default 0x80000000 if VMSPLIT_2G
844 default 0x78000000 if VMSPLIT_2G_OPT
845 default 0x40000000 if VMSPLIT_1G
851 depends on X86_32 && (HIGHMEM64G || HIGHMEM4G)
855 prompt "PAE (Physical Address Extension) Support"
856 depends on X86_32 && !HIGHMEM4G
857 select RESOURCES_64BIT
859 PAE is required for NX support, and furthermore enables
860 larger swapspace support for non-overcommit purposes. It
861 has the cost of more pagetable lookup overhead, and also
862 consumes more pagetable space per process.
864 # Common NUMA Features
866 bool "Numa Memory Allocation and Scheduler Support (EXPERIMENTAL)"
868 depends on X86_64 || (X86_32 && HIGHMEM64G && (X86_NUMAQ || (X86_SUMMIT || X86_GENERICARCH) && ACPI) && EXPERIMENTAL)
870 default y if (X86_NUMAQ || X86_SUMMIT)
872 Enable NUMA (Non Uniform Memory Access) support.
873 The kernel will try to allocate memory used by a CPU on the
874 local memory controller of the CPU and add some more
875 NUMA awareness to the kernel.
877 For i386 this is currently highly experimental and should be only
878 used for kernel development. It might also cause boot failures.
879 For x86_64 this is recommended on all multiprocessor Opteron systems.
880 If the system is EM64T, you should say N unless your system is
883 comment "NUMA (Summit) requires SMP, 64GB highmem support, ACPI"
884 depends on X86_32 && X86_SUMMIT && (!HIGHMEM64G || !ACPI)
888 prompt "Old style AMD Opteron NUMA detection"
889 depends on X86_64 && NUMA && PCI
891 Enable K8 NUMA node topology detection. You should say Y here if
892 you have a multi processor AMD K8 system. This uses an old
893 method to read the NUMA configuration directly from the builtin
894 Northbridge of Opteron. It is recommended to use X86_64_ACPI_NUMA
895 instead, which also takes priority if both are compiled in.
897 config X86_64_ACPI_NUMA
899 prompt "ACPI NUMA detection"
900 depends on X86_64 && NUMA && ACPI && PCI
903 Enable ACPI SRAT based node topology detection.
906 bool "NUMA emulation"
907 depends on X86_64 && NUMA
909 Enable NUMA emulation. A flat machine will be split
910 into virtual nodes when booted with "numa=fake=N", where N is the
911 number of nodes. This is only useful for debugging.
914 int "Max num nodes shift(1-15)"
916 default "6" if X86_64
917 default "4" if X86_NUMAQ
919 depends on NEED_MULTIPLE_NODES
921 config HAVE_ARCH_BOOTMEM_NODE
923 depends on X86_32 && NUMA
925 config ARCH_HAVE_MEMORY_PRESENT
927 depends on X86_32 && DISCONTIGMEM
929 config NEED_NODE_MEMMAP_SIZE
931 depends on X86_32 && (DISCONTIGMEM || SPARSEMEM)
933 config HAVE_ARCH_ALLOC_REMAP
935 depends on X86_32 && NUMA
937 config ARCH_FLATMEM_ENABLE
939 depends on X86_32 && ARCH_SELECT_MEMORY_MODEL && X86_PC && !NUMA
941 config ARCH_DISCONTIGMEM_ENABLE
943 depends on NUMA && X86_32
945 config ARCH_DISCONTIGMEM_DEFAULT
947 depends on NUMA && X86_32
949 config ARCH_SPARSEMEM_DEFAULT
953 config ARCH_SPARSEMEM_ENABLE
955 depends on X86_64 || NUMA || (EXPERIMENTAL && X86_PC)
956 select SPARSEMEM_STATIC if X86_32
957 select SPARSEMEM_VMEMMAP_ENABLE if X86_64
959 config ARCH_SELECT_MEMORY_MODEL
961 depends on ARCH_SPARSEMEM_ENABLE
963 config ARCH_MEMORY_PROBE
965 depends on MEMORY_HOTPLUG
970 bool "Allocate 3rd-level pagetables from highmem"
971 depends on X86_32 && (HIGHMEM4G || HIGHMEM64G)
973 The VM uses one page table entry for each page of physical memory.
974 For systems with a lot of RAM, this can be wasteful of precious
975 low memory. Setting this option will put user-space page table
976 entries in high memory.
978 config MATH_EMULATION
980 prompt "Math emulation" if X86_32
982 Linux can emulate a math coprocessor (used for floating point
983 operations) if you don't have one. 486DX and Pentium processors have
984 a math coprocessor built in, 486SX and 386 do not, unless you added
985 a 487DX or 387, respectively. (The messages during boot time can
986 give you some hints here ["man dmesg"].) Everyone needs either a
987 coprocessor or this emulation.
989 If you don't have a math coprocessor, you need to say Y here; if you
990 say Y here even though you have a coprocessor, the coprocessor will
991 be used nevertheless. (This behavior can be changed with the kernel
992 command line option "no387", which comes handy if your coprocessor
993 is broken. Try "man bootparam" or see the documentation of your boot
994 loader (lilo or loadlin) about how to pass options to the kernel at
995 boot time.) This means that it is a good idea to say Y here if you
996 intend to use this kernel on different machines.
998 More information about the internals of the Linux math coprocessor
999 emulation can be found in <file:arch/x86/math-emu/README>.
1001 If you are not sure, say Y; apart from resulting in a 66 KB bigger
1002 kernel, it won't hurt.
1005 bool "MTRR (Memory Type Range Register) support"
1007 On Intel P6 family processors (Pentium Pro, Pentium II and later)
1008 the Memory Type Range Registers (MTRRs) may be used to control
1009 processor access to memory ranges. This is most useful if you have
1010 a video (VGA) card on a PCI or AGP bus. Enabling write-combining
1011 allows bus write transfers to be combined into a larger transfer
1012 before bursting over the PCI/AGP bus. This can increase performance
1013 of image write operations 2.5 times or more. Saying Y here creates a
1014 /proc/mtrr file which may be used to manipulate your processor's
1015 MTRRs. Typically the X server should use this.
1017 This code has a reasonably generic interface so that similar
1018 control registers on other processors can be easily supported
1021 The Cyrix 6x86, 6x86MX and M II processors have Address Range
1022 Registers (ARRs) which provide a similar functionality to MTRRs. For
1023 these, the ARRs are used to emulate the MTRRs.
1024 The AMD K6-2 (stepping 8 and above) and K6-3 processors have two
1025 MTRRs. The Centaur C6 (WinChip) has 8 MCRs, allowing
1026 write-combining. All of these processors are supported by this code
1027 and it makes sense to say Y here if you have one of them.
1029 Saying Y here also fixes a problem with buggy SMP BIOSes which only
1030 set the MTRRs for the boot CPU and not for the secondary CPUs. This
1031 can lead to all sorts of problems, so it's good to say Y here.
1033 You can safely say Y even if your machine doesn't have MTRRs, you'll
1034 just add about 9 KB to your kernel.
1036 See <file:Documentation/mtrr.txt> for more information.
1040 prompt "x86 PAT support"
1041 depends on MTRR && NONPROMISC_DEVMEM
1043 Use PAT attributes to setup page level cache control.
1045 PATs are the modern equivalents of MTRRs and are much more
1046 flexible than MTRRs.
1048 Say N here if you see bootup problems (boot crash, boot hang,
1049 spontaneous reboots) or a non-working video driver.
1055 prompt "EFI runtime service support"
1058 This enables the kernel to use EFI runtime services that are
1059 available (such as the EFI variable services).
1061 This option is only useful on systems that have EFI firmware.
1062 In addition, you should use the latest ELILO loader available
1063 at <http://elilo.sourceforge.net> in order to take advantage
1064 of EFI runtime services. However, even with this option, the
1065 resultant kernel should continue to boot on existing non-EFI
1070 prompt "Enable kernel irq balancing"
1071 depends on X86_32 && SMP && X86_IO_APIC
1073 The default yes will allow the kernel to do irq load balancing.
1074 Saying no will keep the kernel from doing irq load balancing.
1078 prompt "Enable seccomp to safely compute untrusted bytecode"
1081 This kernel feature is useful for number crunching applications
1082 that may need to compute untrusted bytecode during their
1083 execution. By using pipes or other transports made available to
1084 the process as file descriptors supporting the read/write
1085 syscalls, it's possible to isolate those applications in
1086 their own address space using seccomp. Once seccomp is
1087 enabled via /proc/<pid>/seccomp, it cannot be disabled
1088 and the task is only allowed to execute a few safe syscalls
1089 defined by each seccomp mode.
1091 If unsure, say Y. Only embedded should say N here.
1093 config CC_STACKPROTECTOR
1094 bool "Enable -fstack-protector buffer overflow detection (EXPERIMENTAL)"
1095 depends on X86_64 && EXPERIMENTAL && BROKEN
1097 This option turns on the -fstack-protector GCC feature. This
1098 feature puts, at the beginning of critical functions, a canary
1099 value on the stack just before the return address, and validates
1100 the value just before actually returning. Stack based buffer
1101 overflows (that need to overwrite this return address) now also
1102 overwrite the canary, which gets detected and the attack is then
1103 neutralized via a kernel panic.
1105 This feature requires gcc version 4.2 or above, or a distribution
1106 gcc with the feature backported. Older versions are automatically
1107 detected and for those versions, this configuration option is ignored.
1109 config CC_STACKPROTECTOR_ALL
1110 bool "Use stack-protector for all functions"
1111 depends on CC_STACKPROTECTOR
1113 Normally, GCC only inserts the canary value protection for
1114 functions that use large-ish on-stack buffers. By enabling
1115 this option, GCC will be asked to do this for ALL functions.
1117 source kernel/Kconfig.hz
1120 bool "kexec system call"
1121 depends on X86_64 || X86_BIOS_REBOOT
1123 kexec is a system call that implements the ability to shutdown your
1124 current kernel, and to start another kernel. It is like a reboot
1125 but it is independent of the system firmware. And like a reboot
1126 you can start any kernel with it, not just Linux.
1128 The name comes from the similarity to the exec system call.
1130 It is an ongoing process to be certain the hardware in a machine
1131 is properly shutdown, so do not be surprised if this code does not
1132 initially work for you. It may help to enable device hotplugging
1133 support. As of this writing the exact hardware interface is
1134 strongly in flux, so no good recommendation can be made.
1137 bool "kernel crash dumps (EXPERIMENTAL)"
1138 depends on EXPERIMENTAL
1139 depends on X86_64 || (X86_32 && HIGHMEM)
1141 Generate crash dump after being started by kexec.
1142 This should be normally only set in special crash dump kernels
1143 which are loaded in the main kernel with kexec-tools into
1144 a specially reserved region and then later executed after
1145 a crash by kdump/kexec. The crash dump kernel must be compiled
1146 to a memory address not used by the main kernel or BIOS using
1147 PHYSICAL_START, or it must be built as a relocatable image
1148 (CONFIG_RELOCATABLE=y).
1149 For more details see Documentation/kdump/kdump.txt
1151 config PHYSICAL_START
1152 hex "Physical address where the kernel is loaded" if (EMBEDDED || CRASH_DUMP)
1153 default "0x1000000" if X86_NUMAQ
1154 default "0x200000" if X86_64
1157 This gives the physical address where the kernel is loaded.
1159 If kernel is a not relocatable (CONFIG_RELOCATABLE=n) then
1160 bzImage will decompress itself to above physical address and
1161 run from there. Otherwise, bzImage will run from the address where
1162 it has been loaded by the boot loader and will ignore above physical
1165 In normal kdump cases one does not have to set/change this option
1166 as now bzImage can be compiled as a completely relocatable image
1167 (CONFIG_RELOCATABLE=y) and be used to load and run from a different
1168 address. This option is mainly useful for the folks who don't want
1169 to use a bzImage for capturing the crash dump and want to use a
1170 vmlinux instead. vmlinux is not relocatable hence a kernel needs
1171 to be specifically compiled to run from a specific memory area
1172 (normally a reserved region) and this option comes handy.
1174 So if you are using bzImage for capturing the crash dump, leave
1175 the value here unchanged to 0x100000 and set CONFIG_RELOCATABLE=y.
1176 Otherwise if you plan to use vmlinux for capturing the crash dump
1177 change this value to start of the reserved region (Typically 16MB
1178 0x1000000). In other words, it can be set based on the "X" value as
1179 specified in the "crashkernel=YM@XM" command line boot parameter
1180 passed to the panic-ed kernel. Typically this parameter is set as
1181 crashkernel=64M@16M. Please take a look at
1182 Documentation/kdump/kdump.txt for more details about crash dumps.
1184 Usage of bzImage for capturing the crash dump is recommended as
1185 one does not have to build two kernels. Same kernel can be used
1186 as production kernel and capture kernel. Above option should have
1187 gone away after relocatable bzImage support is introduced. But it
1188 is present because there are users out there who continue to use
1189 vmlinux for dump capture. This option should go away down the
1192 Don't change this unless you know what you are doing.
1195 bool "Build a relocatable kernel (EXPERIMENTAL)"
1196 depends on EXPERIMENTAL
1198 This builds a kernel image that retains relocation information
1199 so it can be loaded someplace besides the default 1MB.
1200 The relocations tend to make the kernel binary about 10% larger,
1201 but are discarded at runtime.
1203 One use is for the kexec on panic case where the recovery kernel
1204 must live at a different physical address than the primary
1207 Note: If CONFIG_RELOCATABLE=y, then the kernel runs from the address
1208 it has been loaded at and the compile time physical address
1209 (CONFIG_PHYSICAL_START) is ignored.
1211 config PHYSICAL_ALIGN
1213 prompt "Alignment value to which kernel should be aligned" if X86_32
1214 default "0x100000" if X86_32
1215 default "0x200000" if X86_64
1216 range 0x2000 0x400000
1218 This value puts the alignment restrictions on physical address
1219 where kernel is loaded and run from. Kernel is compiled for an
1220 address which meets above alignment restriction.
1222 If bootloader loads the kernel at a non-aligned address and
1223 CONFIG_RELOCATABLE is set, kernel will move itself to nearest
1224 address aligned to above value and run from there.
1226 If bootloader loads the kernel at a non-aligned address and
1227 CONFIG_RELOCATABLE is not set, kernel will ignore the run time
1228 load address and decompress itself to the address it has been
1229 compiled for and run from there. The address for which kernel is
1230 compiled already meets above alignment restrictions. Hence the
1231 end result is that kernel runs from a physical address meeting
1232 above alignment restrictions.
1234 Don't change this unless you know what you are doing.
1237 bool "Support for suspend on SMP and hot-pluggable CPUs (EXPERIMENTAL)"
1238 depends on SMP && HOTPLUG && EXPERIMENTAL && !X86_VOYAGER
1240 Say Y here to experiment with turning CPUs off and on, and to
1241 enable suspend on SMP systems. CPUs can be controlled through
1242 /sys/devices/system/cpu.
1243 Say N if you want to disable CPU hotplug and don't need to
1248 prompt "Compat VDSO support"
1249 depends on X86_32 || IA32_EMULATION
1251 Map the 32-bit VDSO to the predictable old-style address too.
1253 Say N here if you are running a sufficiently recent glibc
1254 version (2.3.3 or later), to remove the high-mapped
1255 VDSO mapping and to exclusively use the randomized VDSO.
1261 config ARCH_ENABLE_MEMORY_HOTPLUG
1263 depends on X86_64 || (X86_32 && HIGHMEM)
1265 config HAVE_ARCH_EARLY_PFN_TO_NID
1269 menu "Power management options"
1270 depends on !X86_VOYAGER
1272 config ARCH_HIBERNATION_HEADER
1274 depends on X86_64 && HIBERNATION
1276 source "kernel/power/Kconfig"
1278 source "drivers/acpi/Kconfig"
1283 depends on APM || APM_MODULE
1286 tristate "APM (Advanced Power Management) BIOS support"
1287 depends on X86_32 && PM_SLEEP && !X86_VISWS
1289 APM is a BIOS specification for saving power using several different
1290 techniques. This is mostly useful for battery powered laptops with
1291 APM compliant BIOSes. If you say Y here, the system time will be
1292 reset after a RESUME operation, the /proc/apm device will provide
1293 battery status information, and user-space programs will receive
1294 notification of APM "events" (e.g. battery status change).
1296 If you select "Y" here, you can disable actual use of the APM
1297 BIOS by passing the "apm=off" option to the kernel at boot time.
1299 Note that the APM support is almost completely disabled for
1300 machines with more than one CPU.
1302 In order to use APM, you will need supporting software. For location
1303 and more information, read <file:Documentation/power/pm.txt> and the
1304 Battery Powered Linux mini-HOWTO, available from
1305 <http://www.tldp.org/docs.html#howto>.
1307 This driver does not spin down disk drives (see the hdparm(8)
1308 manpage ("man 8 hdparm") for that), and it doesn't turn off
1309 VESA-compliant "green" monitors.
1311 This driver does not support the TI 4000M TravelMate and the ACER
1312 486/DX4/75 because they don't have compliant BIOSes. Many "green"
1313 desktop machines also don't have compliant BIOSes, and this driver
1314 may cause those machines to panic during the boot phase.
1316 Generally, if you don't have a battery in your machine, there isn't
1317 much point in using this driver and you should say N. If you get
1318 random kernel OOPSes or reboots that don't seem to be related to
1319 anything, try disabling/enabling this option (or disabling/enabling
1322 Some other things you should try when experiencing seemingly random,
1325 1) make sure that you have enough swap space and that it is
1327 2) pass the "no-hlt" option to the kernel
1328 3) switch on floating point emulation in the kernel and pass
1329 the "no387" option to the kernel
1330 4) pass the "floppy=nodma" option to the kernel
1331 5) pass the "mem=4M" option to the kernel (thereby disabling
1332 all but the first 4 MB of RAM)
1333 6) make sure that the CPU is not over clocked.
1334 7) read the sig11 FAQ at <http://www.bitwizard.nl/sig11/>
1335 8) disable the cache from your BIOS settings
1336 9) install a fan for the video card or exchange video RAM
1337 10) install a better fan for the CPU
1338 11) exchange RAM chips
1339 12) exchange the motherboard.
1341 To compile this driver as a module, choose M here: the
1342 module will be called apm.
1346 config APM_IGNORE_USER_SUSPEND
1347 bool "Ignore USER SUSPEND"
1349 This option will ignore USER SUSPEND requests. On machines with a
1350 compliant APM BIOS, you want to say N. However, on the NEC Versa M
1351 series notebooks, it is necessary to say Y because of a BIOS bug.
1353 config APM_DO_ENABLE
1354 bool "Enable PM at boot time"
1356 Enable APM features at boot time. From page 36 of the APM BIOS
1357 specification: "When disabled, the APM BIOS does not automatically
1358 power manage devices, enter the Standby State, enter the Suspend
1359 State, or take power saving steps in response to CPU Idle calls."
1360 This driver will make CPU Idle calls when Linux is idle (unless this
1361 feature is turned off -- see "Do CPU IDLE calls", below). This
1362 should always save battery power, but more complicated APM features
1363 will be dependent on your BIOS implementation. You may need to turn
1364 this option off if your computer hangs at boot time when using APM
1365 support, or if it beeps continuously instead of suspending. Turn
1366 this off if you have a NEC UltraLite Versa 33/C or a Toshiba
1367 T400CDT. This is off by default since most machines do fine without
1371 bool "Make CPU Idle calls when idle"
1373 Enable calls to APM CPU Idle/CPU Busy inside the kernel's idle loop.
1374 On some machines, this can activate improved power savings, such as
1375 a slowed CPU clock rate, when the machine is idle. These idle calls
1376 are made after the idle loop has run for some length of time (e.g.,
1377 333 mS). On some machines, this will cause a hang at boot time or
1378 whenever the CPU becomes idle. (On machines with more than one CPU,
1379 this option does nothing.)
1381 config APM_DISPLAY_BLANK
1382 bool "Enable console blanking using APM"
1384 Enable console blanking using the APM. Some laptops can use this to
1385 turn off the LCD backlight when the screen blanker of the Linux
1386 virtual console blanks the screen. Note that this is only used by
1387 the virtual console screen blanker, and won't turn off the backlight
1388 when using the X Window system. This also doesn't have anything to
1389 do with your VESA-compliant power-saving monitor. Further, this
1390 option doesn't work for all laptops -- it might not turn off your
1391 backlight at all, or it might print a lot of errors to the console,
1392 especially if you are using gpm.
1394 config APM_ALLOW_INTS
1395 bool "Allow interrupts during APM BIOS calls"
1397 Normally we disable external interrupts while we are making calls to
1398 the APM BIOS as a measure to lessen the effects of a badly behaving
1399 BIOS implementation. The BIOS should reenable interrupts if it
1400 needs to. Unfortunately, some BIOSes do not -- especially those in
1401 many of the newer IBM Thinkpads. If you experience hangs when you
1402 suspend, try setting this to Y. Otherwise, say N.
1404 config APM_REAL_MODE_POWER_OFF
1405 bool "Use real mode APM BIOS call to power off"
1407 Use real mode APM BIOS calls to switch off the computer. This is
1408 a work-around for a number of buggy BIOSes. Switch this option on if
1409 your computer crashes instead of powering off properly.
1413 source "arch/x86/kernel/cpu/cpufreq/Kconfig"
1415 source "drivers/cpuidle/Kconfig"
1420 menu "Bus options (PCI etc.)"
1423 bool "PCI support" if !X86_VISWS && !X86_VSMP
1424 depends on !X86_VOYAGER
1426 select ARCH_SUPPORTS_MSI if (X86_LOCAL_APIC && X86_IO_APIC)
1428 Find out whether you have a PCI motherboard. PCI is the name of a
1429 bus system, i.e. the way the CPU talks to the other stuff inside
1430 your box. Other bus systems are ISA, EISA, MicroChannel (MCA) or
1431 VESA. If you have PCI, say Y, otherwise N.
1434 prompt "PCI access mode"
1435 depends on X86_32 && PCI && !X86_VISWS
1438 On PCI systems, the BIOS can be used to detect the PCI devices and
1439 determine their configuration. However, some old PCI motherboards
1440 have BIOS bugs and may crash if this is done. Also, some embedded
1441 PCI-based systems don't have any BIOS at all. Linux can also try to
1442 detect the PCI hardware directly without using the BIOS.
1444 With this option, you can specify how Linux should detect the
1445 PCI devices. If you choose "BIOS", the BIOS will be used,
1446 if you choose "Direct", the BIOS won't be used, and if you
1447 choose "MMConfig", then PCI Express MMCONFIG will be used.
1448 If you choose "Any", the kernel will try MMCONFIG, then the
1449 direct access method and falls back to the BIOS if that doesn't
1450 work. If unsure, go with the default, which is "Any".
1455 config PCI_GOMMCONFIG
1468 depends on X86_32 && !X86_VISWS && PCI && (PCI_GOBIOS || PCI_GOANY)
1470 # x86-64 doesn't support PCI BIOS access from long mode so always go direct.
1473 depends on PCI && (X86_64 || (PCI_GODIRECT || PCI_GOANY) || X86_VISWS)
1477 depends on X86_32 && PCI && ACPI && (PCI_GOMMCONFIG || PCI_GOANY)
1484 bool "Support mmconfig PCI config space access"
1485 depends on X86_64 && PCI && ACPI
1488 bool "Support for DMA Remapping Devices (EXPERIMENTAL)"
1489 depends on X86_64 && PCI_MSI && ACPI && EXPERIMENTAL
1491 DMA remapping (DMAR) devices support enables independent address
1492 translations for Direct Memory Access (DMA) from devices.
1493 These DMA remapping devices are reported via ACPI tables
1494 and include PCI device scope covered by these DMA
1499 prompt "Support for Graphics workaround"
1502 Current Graphics drivers tend to use physical address
1503 for DMA and avoid using DMA APIs. Setting this config
1504 option permits the IOMMU driver to set a unity map for
1505 all the OS-visible memory. Hence the driver can continue
1506 to use physical addresses for DMA.
1508 config DMAR_FLOPPY_WA
1512 Floppy disk drivers are know to bypass DMA API calls
1513 thereby failing to work when IOMMU is enabled. This
1514 workaround will setup a 1:1 mapping for the first
1515 16M to make floppy (an ISA device) work.
1517 source "drivers/pci/pcie/Kconfig"
1519 source "drivers/pci/Kconfig"
1521 # x86_64 have no ISA slots, but do have ISA-style DMA.
1529 depends on !(X86_VOYAGER || X86_VISWS)
1531 Find out whether you have ISA slots on your motherboard. ISA is the
1532 name of a bus system, i.e. the way the CPU talks to the other stuff
1533 inside your box. Other bus systems are PCI, EISA, MicroChannel
1534 (MCA) or VESA. ISA is an older system, now being displaced by PCI;
1535 newer boards don't support it. If you have ISA, say Y, otherwise N.
1541 The Extended Industry Standard Architecture (EISA) bus was
1542 developed as an open alternative to the IBM MicroChannel bus.
1544 The EISA bus provided some of the features of the IBM MicroChannel
1545 bus while maintaining backward compatibility with cards made for
1546 the older ISA bus. The EISA bus saw limited use between 1988 and
1547 1995 when it was made obsolete by the PCI bus.
1549 Say Y here if you are building a kernel for an EISA-based machine.
1553 source "drivers/eisa/Kconfig"
1556 bool "MCA support" if !(X86_VISWS || X86_VOYAGER)
1557 default y if X86_VOYAGER
1559 MicroChannel Architecture is found in some IBM PS/2 machines and
1560 laptops. It is a bus system similar to PCI or ISA. See
1561 <file:Documentation/mca.txt> (and especially the web page given
1562 there) before attempting to build an MCA bus kernel.
1564 source "drivers/mca/Kconfig"
1567 tristate "NatSemi SCx200 support"
1568 depends on !X86_VOYAGER
1570 This provides basic support for National Semiconductor's
1571 (now AMD's) Geode processors. The driver probes for the
1572 PCI-IDs of several on-chip devices, so its a good dependency
1573 for other scx200_* drivers.
1575 If compiled as a module, the driver is named scx200.
1577 config SCx200HR_TIMER
1578 tristate "NatSemi SCx200 27MHz High-Resolution Timer Support"
1579 depends on SCx200 && GENERIC_TIME
1582 This driver provides a clocksource built upon the on-chip
1583 27MHz high-resolution timer. Its also a workaround for
1584 NSC Geode SC-1100's buggy TSC, which loses time when the
1585 processor goes idle (as is done by the scheduler). The
1586 other workaround is idle=poll boot option.
1588 config GEODE_MFGPT_TIMER
1590 prompt "Geode Multi-Function General Purpose Timer (MFGPT) events"
1591 depends on MGEODE_LX && GENERIC_TIME && GENERIC_CLOCKEVENTS
1593 This driver provides a clock event source based on the MFGPT
1594 timer(s) in the CS5535 and CS5536 companion chip for the geode.
1595 MFGPTs have a better resolution and max interval than the
1596 generic PIT, and are suitable for use as high-res timers.
1602 depends on AGP_AMD64 || (X86_64 && (GART_IOMMU || (PCI && NUMA)))
1604 source "drivers/pcmcia/Kconfig"
1606 source "drivers/pci/hotplug/Kconfig"
1611 menu "Executable file formats / Emulations"
1613 source "fs/Kconfig.binfmt"
1615 config IA32_EMULATION
1616 bool "IA32 Emulation"
1618 select COMPAT_BINFMT_ELF
1620 Include code to run 32-bit programs under a 64-bit kernel. You should
1621 likely turn this on, unless you're 100% sure that you don't have any
1622 32-bit programs left.
1625 tristate "IA32 a.out support"
1626 depends on IA32_EMULATION && ARCH_SUPPORTS_AOUT
1628 Support old a.out binaries in the 32bit emulation.
1632 depends on IA32_EMULATION
1634 config COMPAT_FOR_U64_ALIGNMENT
1638 config SYSVIPC_COMPAT
1640 depends on X86_64 && COMPAT && SYSVIPC
1645 source "net/Kconfig"
1647 source "drivers/Kconfig"
1649 source "drivers/firmware/Kconfig"
1653 source "arch/x86/Kconfig.debug"
1655 source "security/Kconfig"
1657 source "crypto/Kconfig"
1659 source "arch/x86/kvm/Kconfig"
1661 source "lib/Kconfig"
projects / deliverable / linux.git / blob