2 # For a description of the syntax of this configuration file,
3 # see Documentation/kbuild/kconfig-language.txt.
6 mainmenu "Linux Kernel Configuration"
12 This is Linux's home port. Linux was originally native to the Intel
13 386, and runs on all the later x86 processors including the Intel
14 486, 586, Pentiums, and various instruction-set-compatible chips by
15 AMD, Cyrix, and others.
28 config GENERIC_ISA_DMA
35 menu "Processor type and features"
38 prompt "Subarchitecture Type"
44 Choose this option if your computer is a standard PC or compatible.
49 Select this for an AMD Elan processor.
51 Do not use this option for K6/Athlon/Opteron processors!
53 If unsure, choose "PC-compatible" instead.
58 Voyager is a MCA based 32 way capable SMP architecture proprietary
59 to NCR Corp. Machine classes 345x/35xx/4100/51xx are voyager based.
63 If you do not specifically know you have a Voyager based machine,
64 say N here otherwise the kernel you build will not be bootable.
67 bool "NUMAQ (IBM/Sequent)"
69 This option is used for getting Linux to run on a (IBM/Sequent) NUMA
70 multiquad box. This changes the way that processors are bootstrapped,
71 and uses Clustered Logical APIC addressing mode instead of Flat Logical.
72 You will need a new lynxer.elf file to flash your firmware with - send
73 email to <Martin.Bligh@us.ibm.com>.
76 bool "Summit/EXA (IBM x440)"
79 This option is needed for IBM systems that use the Summit/EXA chipset.
80 In particular, it is needed for the x440.
82 If you don't have one of these computers, you should say N here.
85 bool "Support for other sub-arch SMP systems with more than 8 CPUs"
88 This option is needed for the systems that have more than 8 CPUs
89 and if the system is not of any sub-arch type above.
91 If you don't have such a system, you should say N here.
94 bool "SGI 320/540 (Visual Workstation)"
96 The SGI Visual Workstation series is an IA32-based workstation
97 based on SGI systems chips with some legacy PC hardware attached.
99 Say Y here to create a kernel to run on the SGI 320 or 540.
101 A kernel compiled for the Visual Workstation will not run on PCs
102 and vice versa. See <file:Documentation/sgi-visws.txt> for details.
104 config X86_GENERICARCH
105 bool "Generic architecture (Summit, bigsmp, ES7000, default)"
108 This option compiles in the Summit, bigsmp, ES7000, default subarchitectures.
109 It is intended for a generic binary kernel.
112 bool "Support for Unisys ES7000 IA32 series"
115 Support for Unisys ES7000 systems. Say 'Y' here if this kernel is
116 supposed to run on an IA32-based Unisys ES7000 system.
117 Only choose this option if you have such a system, otherwise you
125 depends on NUMA && (X86_SUMMIT || X86_GENERICARCH)
127 config X86_SUMMIT_NUMA
130 depends on NUMA && (X86_SUMMIT || X86_GENERICARCH)
132 config X86_CYCLONE_TIMER
135 depends on X86_SUMMIT || X86_GENERICARCH
137 config ES7000_CLUSTERED_APIC
140 depends on SMP && X86_ES7000 && MPENTIUMIII
145 prompt "Processor family"
151 This is the processor type of your CPU. This information is used for
152 optimizing purposes. In order to compile a kernel that can run on
153 all x86 CPU types (albeit not optimally fast), you can specify
156 The kernel will not necessarily run on earlier architectures than
157 the one you have chosen, e.g. a Pentium optimized kernel will run on
158 a PPro, but not necessarily on a i486.
160 Here are the settings recommended for greatest speed:
161 - "386" for the AMD/Cyrix/Intel 386DX/DXL/SL/SLC/SX, Cyrix/TI
162 486DLC/DLC2, UMC 486SX-S and NexGen Nx586. Only "386" kernels
163 will run on a 386 class machine.
164 - "486" for the AMD/Cyrix/IBM/Intel 486DX/DX2/DX4 or
165 SL/SLC/SLC2/SLC3/SX/SX2 and UMC U5D or U5S.
166 - "586" for generic Pentium CPUs lacking the TSC
167 (time stamp counter) register.
168 - "Pentium-Classic" for the Intel Pentium.
169 - "Pentium-MMX" for the Intel Pentium MMX.
170 - "Pentium-Pro" for the Intel Pentium Pro.
171 - "Pentium-II" for the Intel Pentium II or pre-Coppermine Celeron.
172 - "Pentium-III" for the Intel Pentium III or Coppermine Celeron.
173 - "Pentium-4" for the Intel Pentium 4 or P4-based Celeron.
174 - "K6" for the AMD K6, K6-II and K6-III (aka K6-3D).
175 - "Athlon" for the AMD K7 family (Athlon/Duron/Thunderbird).
176 - "Crusoe" for the Transmeta Crusoe series.
177 - "Winchip-C6" for original IDT Winchip.
178 - "Winchip-2" for IDT Winchip 2.
179 - "Winchip-2A" for IDT Winchips with 3dNow! capabilities.
180 - "CyrixIII/VIA C3" for VIA Cyrix III or VIA C3.
181 - "VIA C3-2 for VIA C3-2 "Nehemiah" (model 9 and above).
183 If you don't know what to do, choose "386".
188 Select this for a 486 series processor, either Intel or one of the
189 compatible processors from AMD, Cyrix, IBM, or Intel. Includes DX,
190 DX2, and DX4 variants; also SL/SLC/SLC2/SLC3/SX/SX2 and UMC U5D or
194 bool "586/K5/5x86/6x86/6x86MX"
196 Select this for an 586 or 686 series processor such as the AMD K5,
197 the Intel 5x86 or 6x86, or the Intel 6x86MX. This choice does not
198 assume the RDTSC (Read Time Stamp Counter) instruction.
201 bool "Pentium-Classic"
203 Select this for a Pentium Classic processor with the RDTSC (Read
204 Time Stamp Counter) instruction for benchmarking.
209 Select this for a Pentium with the MMX graphics/multimedia
210 extended instructions.
215 Select this for Intel Pentium Pro chips. This enables the use of
216 Pentium Pro extended instructions, and disables the init-time guard
217 against the f00f bug found in earlier Pentiums.
220 bool "Pentium-II/Celeron(pre-Coppermine)"
222 Select this for Intel chips based on the Pentium-II and
223 pre-Coppermine Celeron core. This option enables an unaligned
224 copy optimization, compiles the kernel with optimization flags
225 tailored for the chip, and applies any applicable Pentium Pro
229 bool "Pentium-III/Celeron(Coppermine)/Pentium-III Xeon"
231 Select this for Intel chips based on the Pentium-III and
232 Celeron-Coppermine core. This option enables use of some
233 extended prefetch instructions in addition to the Pentium II
239 Select this for Intel Pentium M (not Pentium-4 M)
243 bool "Pentium-4/Celeron(P4-based)/Pentium-4 M/Xeon"
245 Select this for Intel Pentium 4 chips. This includes the
246 Pentium 4, P4-based Celeron and Xeon, and Pentium-4 M
247 (not Pentium M) chips. This option enables compile flags
248 optimized for the chip, uses the correct cache shift, and
249 applies any applicable Pentium III optimizations.
252 bool "K6/K6-II/K6-III"
254 Select this for an AMD K6-family processor. Enables use of
255 some extended instructions, and passes appropriate optimization
259 bool "Athlon/Duron/K7"
261 Select this for an AMD Athlon K7-family processor. Enables use of
262 some extended instructions, and passes appropriate optimization
266 bool "Opteron/Athlon64/Hammer/K8"
268 Select this for an AMD Opteron or Athlon64 Hammer-family processor. Enables
269 use of some extended instructions, and passes appropriate optimization
275 Select this for a Transmeta Crusoe processor. Treats the processor
276 like a 586 with TSC, and sets some GCC optimization flags (like a
277 Pentium Pro with no alignment requirements).
282 Select this for an IDT Winchip C6 chip. Linux and GCC
283 treat this chip as a 586TSC with some extended instructions
284 and alignment requirements.
289 Select this for an IDT Winchip-2. Linux and GCC
290 treat this chip as a 586TSC with some extended instructions
291 and alignment requirements.
294 bool "Winchip-2A/Winchip-3"
296 Select this for an IDT Winchip-2A or 3. Linux and GCC
297 treat this chip as a 586TSC with some extended instructions
298 and alignment reqirements. Also enable out of order memory
299 stores for this CPU, which can increase performance of some
303 bool "CyrixIII/VIA-C3"
305 Select this for a Cyrix III or C3 chip. Presently Linux and GCC
306 treat this chip as a generic 586. Whilst the CPU is 686 class,
307 it lacks the cmov extension which gcc assumes is present when
309 Note that Nehemiah (Model 9) and above will not boot with this
310 kernel due to them lacking the 3DNow! instructions used in earlier
311 incarnations of the CPU.
314 bool "VIA C3-2 (Nehemiah)"
316 Select this for a VIA C3 "Nehemiah". Selecting this enables usage
317 of SSE and tells gcc to treat the CPU as a 686.
318 Note, this kernel will not boot on older (pre model 9) C3s.
323 bool "Generic x86 support"
325 Instead of just including optimizations for the selected
326 x86 variant (e.g. PII, Crusoe or Athlon), include some more
327 generic optimizations as well. This will make the kernel
328 perform better on x86 CPUs other than that selected.
330 This is really intended for distributors who need more
331 generic optimizations.
336 # Define implied options from the CPU selection here
348 config X86_L1_CACHE_SHIFT
350 default "7" if MPENTIUM4 || X86_GENERIC
351 default "4" if X86_ELAN || M486 || M386
352 default "5" if MWINCHIP3D || MWINCHIP2 || MWINCHIPC6 || MCRUSOE || MCYRIXIII || MK6 || MPENTIUMIII || MPENTIUMII || M686 || M586MMX || M586TSC || M586 || MVIAC3_2
353 default "6" if MK7 || MK8 || MPENTIUMM
355 config RWSEM_GENERIC_SPINLOCK
360 config RWSEM_XCHGADD_ALGORITHM
365 config X86_PPRO_FENCE
367 depends on M686 || M586MMX || M586TSC || M586 || M486 || M386
372 depends on M586MMX || M586TSC || M586 || M486 || M386
375 config X86_WP_WORKS_OK
395 config X86_ALIGNMENT_16
397 depends on MWINCHIP3D || MWINCHIP2 || MWINCHIPC6 || MCYRIXIII || X86_ELAN || MK6 || M586MMX || M586TSC || M586 || M486 || MVIAC3_2
402 depends on MK7 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || M586MMX || MK8
405 config X86_INTEL_USERCOPY
407 depends on MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M586MMX || X86_GENERIC || MK8 || MK7
410 config X86_USE_PPRO_CHECKSUM
412 depends on MWINCHIP3D || MWINCHIP2 || MWINCHIPC6 || MCYRIXIII || MK7 || MK6 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || MK8 || MVIAC3_2
417 depends on MCYRIXIII || MK7
422 depends on (MWINCHIP3D || MWINCHIP2 || MWINCHIPC6) && MTRR
426 bool "4 GB kernel-space and 4 GB user-space virtual memory support"
428 This option is only useful for systems that have more than 1 GB
431 The default kernel VM layout leaves 1 GB of virtual memory for
432 kernel-space mappings, and 3 GB of VM for user-space applications.
433 This option ups both the kernel-space VM and the user-space VM to
436 The cost of this option is additional TLB flushes done at
437 system-entry points that transition from user-mode into kernel-mode.
438 I.e. system calls and page faults, and IRQs that interrupt user-mode
439 code. There's also additional overhead to kernel operations that copy
440 memory to/from user-space. The overhead from this is hard to tell and
441 depends on the workload - it can be anything from no visible overhead
442 to 20-30% overhead. A good rule of thumb is to count with a runtime
445 The upside is the much increased kernel-space VM, which more than
446 quadruples the maximum amount of RAM supported. Kernels compiled with
447 this option boot on 64GB of RAM and still have more than 3.1 GB of
448 'lowmem' left. Another bonus is that highmem IO bouncing decreases,
449 if used with drivers that still use bounce-buffers.
451 There's also a 33% increase in user-space VM size - database
452 applications might see a boost from this.
454 But the cost of the TLB flushes and the runtime overhead has to be
455 weighed against the bonuses offered by the larger VM spaces. The
456 dividing line depends on the actual workload - there might be 4 GB
457 systems that benefit from this option. Systems with less than 4 GB
458 of RAM will rarely see a benefit from this option - but it's not
459 out of question, the exact circumstances have to be considered.
461 config X86_SWITCH_PAGETABLES
464 config X86_4G_VM_LAYOUT
467 config X86_UACCESS_INDIRECT
470 config X86_HIGH_ENTRY
474 bool "HPET Timer Support"
476 This enables the use of the HPET for the kernel's internal timer.
477 HPET is the next generation timer replacing legacy 8254s.
478 You can safely choose Y here. However, HPET will only be
479 activated if the platform and the BIOS support this feature.
480 Otherwise the 8254 will be used for timing services.
482 Choose N to continue using the legacy 8254 timer.
484 config HPET_EMULATE_RTC
485 def_bool HPET_TIMER && RTC=y
488 bool "Symmetric multi-processing support"
490 This enables support for systems with more than one CPU. If you have
491 a system with only one CPU, like most personal computers, say N. If
492 you have a system with more than one CPU, say Y.
494 If you say N here, the kernel will run on single and multiprocessor
495 machines, but will use only one CPU of a multiprocessor machine. If
496 you say Y here, the kernel will run on many, but not all,
497 singleprocessor machines. On a singleprocessor machine, the kernel
498 will run faster if you say N here.
500 Note that if you say Y here and choose architecture "586" or
501 "Pentium" under "Processor family", the kernel will not work on 486
502 architectures. Similarly, multiprocessor kernels for the "PPro"
503 architecture may not work on all Pentium based boards.
505 People using multiprocessor machines who say Y here should also say
506 Y to "Enhanced Real Time Clock Support", below. The "Advanced Power
507 Management" code will be disabled if you say Y here.
509 See also the <file:Documentation/smp.txt>,
510 <file:Documentation/i386/IO-APIC.txt>,
511 <file:Documentation/nmi_watchdog.txt> and the SMP-HOWTO available at
512 <http://www.tldp.org/docs.html#howto>.
514 If you don't know what to do here, say N.
517 int "Maximum number of CPUs (2-255)"
520 default "32" if X86_NUMAQ || X86_SUMMIT || X86_BIGSMP || X86_ES7000
523 This allows you to specify the maximum number of CPUs which this
524 kernel will support. The maximum supported value is 255 and the
525 minimum value which makes sense is 2.
527 This is purely to save memory - each supported CPU adds
528 approximately eight kilobytes to the kernel image.
531 bool "SMT (Hyperthreading) scheduler support"
535 SMT scheduler support improves the CPU scheduler's decision making
536 when dealing with Intel Pentium 4 chips with HyperThreading at a
537 cost of slightly increased overhead in some places. If unsure say
541 bool "Preemptible Kernel"
543 This option reduces the latency of the kernel when reacting to
544 real-time or interactive events by allowing a low priority process to
545 be preempted even if it is in kernel mode executing a system call.
546 This allows applications to run more reliably even when the system is
549 Say Y here if you are building a kernel for a desktop, embedded
550 or real-time system. Say N if you are unsure.
553 bool "Local APIC support on uniprocessors" if !SMP
554 depends on !(X86_VISWS || X86_VOYAGER)
556 A local APIC (Advanced Programmable Interrupt Controller) is an
557 integrated interrupt controller in the CPU. If you have a single-CPU
558 system which has a processor with a local APIC, you can say Y here to
559 enable and use it. If you say Y here even though your machine doesn't
560 have a local APIC, then the kernel will still run with no slowdown at
561 all. The local APIC supports CPU-generated self-interrupts (timer,
562 performance counters), and the NMI watchdog which detects hard
565 If you have a system with several CPUs, you do not need to say Y
566 here: the local APIC will be used automatically.
569 bool "IO-APIC support on uniprocessors"
570 depends on !SMP && X86_UP_APIC
572 An IO-APIC (I/O Advanced Programmable Interrupt Controller) is an
573 SMP-capable replacement for PC-style interrupt controllers. Most
574 SMP systems and a small number of 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 If you have a system with several CPUs, you do not need to say Y
580 here: the IO-APIC will be used automatically.
582 config X86_LOCAL_APIC
584 depends on !SMP && X86_UP_APIC
589 depends on !SMP && X86_UP_IOAPIC
594 depends on (MWINCHIP3D || MWINCHIP2 || MCRUSOE || MCYRIXIII || MK7 || MK6 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || M586MMX || M586TSC || MK8 || MVIAC3_2) && !X86_NUMAQ
598 bool "Machine Check Exception"
600 Machine Check Exception support allows the processor to notify the
601 kernel if it detects a problem (e.g. overheating, component failure).
602 The action the kernel takes depends on the severity of the problem,
603 ranging from a warning message on the console, to halting the machine.
604 Your processor must be a Pentium or newer to support this - check the
605 flags in /proc/cpuinfo for mce. Note that some older Pentium systems
606 have a design flaw which leads to false MCE events - hence MCE is
607 disabled on all P5 processors, unless explicitly enabled with "mce"
608 as a boot argument. Similarly, if MCE is built in and creates a
609 problem on some new non-standard machine, you can boot with "nomce"
610 to disable it. MCE support simply ignores non-MCE processors like
611 the 386 and 486, so nearly everyone can say Y here.
613 config X86_MCE_NONFATAL
614 tristate "Check for non-fatal errors on AMD Athlon/Duron / Intel Pentium 4"
617 Enabling this feature starts a timer that triggers every 5 seconds which
618 will look at the machine check registers to see if anything happened.
619 Non-fatal problems automatically get corrected (but still logged).
620 Disable this if you don't want to see these messages.
621 Seeing the messages this option prints out may be indicative of dying hardware,
622 or out-of-spec (ie, overclocked) hardware.
623 This option only does something on certain CPUs.
624 (AMD Athlon/Duron and Intel Pentium 4)
626 config X86_MCE_P4THERMAL
627 bool "check for P4 thermal throttling interrupt."
628 depends on X86_MCE && (X86_UP_APIC || SMP)
630 Enabling this feature will cause a message to be printed when the P4
631 enters thermal throttling.
634 tristate "Toshiba Laptop support"
636 This adds a driver to safely access the System Management Mode of
637 the CPU on Toshiba portables with a genuine Toshiba BIOS. It does
638 not work on models with a Phoenix BIOS. The System Management Mode
639 is used to set the BIOS and power saving options on Toshiba portables.
641 For information on utilities to make use of this driver see the
642 Toshiba Linux utilities web site at:
643 <http://www.buzzard.org.uk/toshiba/>.
645 Say Y if you intend to run this kernel on a Toshiba portable.
649 tristate "Dell laptop support"
651 This adds a driver to safely access the System Management Mode
652 of the CPU on the Dell Inspiron 8000. The System Management Mode
653 is used to read cpu temperature and cooling fan status and to
654 control the fans on the I8K portables.
656 This driver has been tested only on the Inspiron 8000 but it may
657 also work with other Dell laptops. You can force loading on other
658 models by passing the parameter `force=1' to the module. Use at
661 For information on utilities to make use of this driver see the
662 I8K Linux utilities web site at:
663 <http://people.debian.org/~dz/i8k/>
665 Say Y if you intend to run this kernel on a Dell Inspiron 8000.
669 tristate "/dev/cpu/microcode - Intel IA32 CPU microcode support"
671 If you say Y here and also to "/dev file system support" in the
672 'File systems' section, you will be able to update the microcode on
673 Intel processors in the IA32 family, e.g. Pentium Pro, Pentium II,
674 Pentium III, Pentium 4, Xeon etc. You will obviously need the
675 actual microcode binary data itself which is not shipped with the
678 For latest news and information on obtaining all the required
679 ingredients for this driver, check:
680 <http://www.urbanmyth.org/microcode/>.
682 To compile this driver as a module, choose M here: the
683 module will be called microcode.
686 tristate "/dev/cpu/*/msr - Model-specific register support"
688 This device gives privileged processes access to the x86
689 Model-Specific Registers (MSRs). It is a character device with
690 major 202 and minors 0 to 31 for /dev/cpu/0/msr to /dev/cpu/31/msr.
691 MSR accesses are directed to a specific CPU on multi-processor
695 tristate "/dev/cpu/*/cpuid - CPU information support"
697 This device gives processes access to the x86 CPUID instruction to
698 be executed on a specific processor. It is a character device
699 with major 203 and minors 0 to 31 for /dev/cpu/0/cpuid to
702 source "drivers/firmware/Kconfig"
705 prompt "High Memory Support"
711 Linux can use up to 64 Gigabytes of physical memory on x86 systems.
712 However, the address space of 32-bit x86 processors is only 4
713 Gigabytes large. That means that, if you have a large amount of
714 physical memory, not all of it can be "permanently mapped" by the
715 kernel. The physical memory that's not permanently mapped is called
718 If you are compiling a kernel which will never run on a machine with
719 more than 1 Gigabyte total physical RAM, answer "off" here (default
720 choice and suitable for most users). This will result in a "3GB/1GB"
721 split: 3GB are mapped so that each process sees a 3GB virtual memory
722 space and the remaining part of the 4GB virtual memory space is used
723 by the kernel to permanently map as much physical memory as
726 If the machine has between 1 and 4 Gigabytes physical RAM, then
729 If more than 4 Gigabytes is used then answer "64GB" here. This
730 selection turns Intel PAE (Physical Address Extension) mode on.
731 PAE implements 3-level paging on IA32 processors. PAE is fully
732 supported by Linux, PAE mode is implemented on all recent Intel
733 processors (Pentium Pro and better). NOTE: If you say "64GB" here,
734 then the kernel will not boot on CPUs that don't support PAE!
736 The actual amount of total physical memory will either be
737 auto detected or can be forced by using a kernel command line option
738 such as "mem=256M". (Try "man bootparam" or see the documentation of
739 your boot loader (lilo or loadlin) about how to pass options to the
740 kernel at boot time.)
742 If unsure, say "off".
747 Select this if you have a 32-bit processor and between 1 and 4
748 gigabytes of physical RAM.
753 Select this if you have a 32-bit processor and more than 4
754 gigabytes of physical RAM.
760 depends on HIGHMEM64G || HIGHMEM4G
765 depends on HIGHMEM64G
768 # Common NUMA Features
770 bool "Numa Memory Allocation and Scheduler Support"
771 depends on SMP && HIGHMEM64G && (X86_NUMAQ || X86_GENERICARCH || (X86_SUMMIT && ACPI))
773 default y if (X86_NUMAQ || X86_SUMMIT)
775 # Need comments to help the hapless user trying to turn on NUMA support
776 comment "NUMA (NUMA-Q) requires SMP, 64GB highmem support"
777 depends on X86_NUMAQ && (!HIGHMEM64G || !SMP)
779 comment "NUMA (Summit) requires SMP, 64GB highmem support, ACPI"
780 depends on X86_SUMMIT && (!HIGHMEM64G || !ACPI)
787 config HAVE_ARCH_BOOTMEM_NODE
793 bool "Allocate 3rd-level pagetables from highmem"
794 depends on HIGHMEM4G || HIGHMEM64G
796 The VM uses one page table entry for each page of physical memory.
797 For systems with a lot of RAM, this can be wasteful of precious
798 low memory. Setting this option will put user-space page table
799 entries in high memory.
801 config MATH_EMULATION
802 bool "Math emulation"
804 Linux can emulate a math coprocessor (used for floating point
805 operations) if you don't have one. 486DX and Pentium processors have
806 a math coprocessor built in, 486SX and 386 do not, unless you added
807 a 487DX or 387, respectively. (The messages during boot time can
808 give you some hints here ["man dmesg"].) Everyone needs either a
809 coprocessor or this emulation.
811 If you don't have a math coprocessor, you need to say Y here; if you
812 say Y here even though you have a coprocessor, the coprocessor will
813 be used nevertheless. (This behavior can be changed with the kernel
814 command line option "no387", which comes handy if your coprocessor
815 is broken. Try "man bootparam" or see the documentation of your boot
816 loader (lilo or loadlin) about how to pass options to the kernel at
817 boot time.) This means that it is a good idea to say Y here if you
818 intend to use this kernel on different machines.
820 More information about the internals of the Linux math coprocessor
821 emulation can be found in <file:arch/i386/math-emu/README>.
823 If you are not sure, say Y; apart from resulting in a 66 KB bigger
824 kernel, it won't hurt.
827 bool "MTRR (Memory Type Range Register) support"
829 On Intel P6 family processors (Pentium Pro, Pentium II and later)
830 the Memory Type Range Registers (MTRRs) may be used to control
831 processor access to memory ranges. This is most useful if you have
832 a video (VGA) card on a PCI or AGP bus. Enabling write-combining
833 allows bus write transfers to be combined into a larger transfer
834 before bursting over the PCI/AGP bus. This can increase performance
835 of image write operations 2.5 times or more. Saying Y here creates a
836 /proc/mtrr file which may be used to manipulate your processor's
837 MTRRs. Typically the X server should use this.
839 This code has a reasonably generic interface so that similar
840 control registers on other processors can be easily supported
843 The Cyrix 6x86, 6x86MX and M II processors have Address Range
844 Registers (ARRs) which provide a similar functionality to MTRRs. For
845 these, the ARRs are used to emulate the MTRRs.
846 The AMD K6-2 (stepping 8 and above) and K6-3 processors have two
847 MTRRs. The Centaur C6 (WinChip) has 8 MCRs, allowing
848 write-combining. All of these processors are supported by this code
849 and it makes sense to say Y here if you have one of them.
851 Saying Y here also fixes a problem with buggy SMP BIOSes which only
852 set the MTRRs for the boot CPU and not for the secondary CPUs. This
853 can lead to all sorts of problems, so it's good to say Y here.
855 You can safely say Y even if your machine doesn't have MTRRs, you'll
856 just add about 9 KB to your kernel.
858 See <file:Documentation/mtrr.txt> for more information.
861 bool "Boot from EFI support (EXPERIMENTAL)"
866 This enables the the kernel to boot on EFI platforms using
867 system configuration information passed to it from the firmware.
868 This also enables the kernel to use any EFI runtime services that are
869 available (such as the EFI variable services).
871 This option is only useful on systems that have EFI firmware
872 and will result in a kernel image that is ~8k larger. In addition,
873 you must use the latest ELILO loader available at
874 <ftp://ftp.hpl.hp.com/pub/linux-ia64/> in order to take advantage of
875 kernel initialization using EFI information (neither GRUB nor LILO know
876 anything about EFI). However, even with this option, the resultant
877 kernel should continue to boot on existing non-EFI platforms.
880 bool "Enable kernel irq balancing"
881 depends on SMP && X86_IO_APIC
884 The default yes will allow the kernel to do irq load balancing.
885 Saying no will keep the kernel from doing irq load balancing.
889 depends on (SMP || PREEMPT) && X86_CMPXCHG
892 # turning this on wastes a bunch of space.
893 # Summit needs it only when NUMA is on
896 depends on (((X86_SUMMIT || X86_GENERICARCH) && NUMA) || (X86 && EFI))
900 bool "Use register arguments (EXPERIMENTAL)"
901 depends on EXPERIMENTAL
904 Compile the kernel with -mregparm=3. This uses an different ABI
905 and passes the first three arguments of a function call in registers.
906 This will probably break binary only modules.
908 This feature is only enabled for gcc-3.0 and later - earlier compilers
909 generate incorrect output with certain kernel constructs when
915 menu "Power management options (ACPI, APM)"
916 depends on !X86_VOYAGER
918 source kernel/power/Kconfig
920 source "drivers/acpi/Kconfig"
922 menu "APM (Advanced Power Management) BIOS Support"
926 tristate "APM (Advanced Power Management) BIOS support"
929 APM is a BIOS specification for saving power using several different
930 techniques. This is mostly useful for battery powered laptops with
931 APM compliant BIOSes. If you say Y here, the system time will be
932 reset after a RESUME operation, the /proc/apm device will provide
933 battery status information, and user-space programs will receive
934 notification of APM "events" (e.g. battery status change).
936 If you select "Y" here, you can disable actual use of the APM
937 BIOS by passing the "apm=off" option to the kernel at boot time.
939 Note that the APM support is almost completely disabled for
940 machines with more than one CPU.
942 In order to use APM, you will need supporting software. For location
943 and more information, read <file:Documentation/pm.txt> and the
944 Battery Powered Linux mini-HOWTO, available from
945 <http://www.tldp.org/docs.html#howto>.
947 This driver does not spin down disk drives (see the hdparm(8)
948 manpage ("man 8 hdparm") for that), and it doesn't turn off
949 VESA-compliant "green" monitors.
951 This driver does not support the TI 4000M TravelMate and the ACER
952 486/DX4/75 because they don't have compliant BIOSes. Many "green"
953 desktop machines also don't have compliant BIOSes, and this driver
954 may cause those machines to panic during the boot phase.
956 Generally, if you don't have a battery in your machine, there isn't
957 much point in using this driver and you should say N. If you get
958 random kernel OOPSes or reboots that don't seem to be related to
959 anything, try disabling/enabling this option (or disabling/enabling
962 Some other things you should try when experiencing seemingly random,
965 1) make sure that you have enough swap space and that it is
967 2) pass the "no-hlt" option to the kernel
968 3) switch on floating point emulation in the kernel and pass
969 the "no387" option to the kernel
970 4) pass the "floppy=nodma" option to the kernel
971 5) pass the "mem=4M" option to the kernel (thereby disabling
972 all but the first 4 MB of RAM)
973 6) make sure that the CPU is not over clocked.
974 7) read the sig11 FAQ at <http://www.bitwizard.nl/sig11/>
975 8) disable the cache from your BIOS settings
976 9) install a fan for the video card or exchange video RAM
977 10) install a better fan for the CPU
978 11) exchange RAM chips
979 12) exchange the motherboard.
981 To compile this driver as a module, choose M here: the
982 module will be called apm.
984 config APM_IGNORE_USER_SUSPEND
985 bool "Ignore USER SUSPEND"
988 This option will ignore USER SUSPEND requests. On machines with a
989 compliant APM BIOS, you want to say N. However, on the NEC Versa M
990 series notebooks, it is necessary to say Y because of a BIOS bug.
993 bool "Enable PM at boot time"
996 Enable APM features at boot time. From page 36 of the APM BIOS
997 specification: "When disabled, the APM BIOS does not automatically
998 power manage devices, enter the Standby State, enter the Suspend
999 State, or take power saving steps in response to CPU Idle calls."
1000 This driver will make CPU Idle calls when Linux is idle (unless this
1001 feature is turned off -- see "Do CPU IDLE calls", below). This
1002 should always save battery power, but more complicated APM features
1003 will be dependent on your BIOS implementation. You may need to turn
1004 this option off if your computer hangs at boot time when using APM
1005 support, or if it beeps continuously instead of suspending. Turn
1006 this off if you have a NEC UltraLite Versa 33/C or a Toshiba
1007 T400CDT. This is off by default since most machines do fine without
1011 bool "Make CPU Idle calls when idle"
1014 Enable calls to APM CPU Idle/CPU Busy inside the kernel's idle loop.
1015 On some machines, this can activate improved power savings, such as
1016 a slowed CPU clock rate, when the machine is idle. These idle calls
1017 are made after the idle loop has run for some length of time (e.g.,
1018 333 mS). On some machines, this will cause a hang at boot time or
1019 whenever the CPU becomes idle. (On machines with more than one CPU,
1020 this option does nothing.)
1022 config APM_DISPLAY_BLANK
1023 bool "Enable console blanking using APM"
1026 Enable console blanking using the APM. Some laptops can use this to
1027 turn off the LCD backlight when the screen blanker of the Linux
1028 virtual console blanks the screen. Note that this is only used by
1029 the virtual console screen blanker, and won't turn off the backlight
1030 when using the X Window system. This also doesn't have anything to
1031 do with your VESA-compliant power-saving monitor. Further, this
1032 option doesn't work for all laptops -- it might not turn off your
1033 backlight at all, or it might print a lot of errors to the console,
1034 especially if you are using gpm.
1036 config APM_RTC_IS_GMT
1037 bool "RTC stores time in GMT"
1040 Say Y here if your RTC (Real Time Clock a.k.a. hardware clock)
1041 stores the time in GMT (Greenwich Mean Time). Say N if your RTC
1044 It is in fact recommended to store GMT in your RTC, because then you
1045 don't have to worry about daylight savings time changes. The only
1046 reason not to use GMT in your RTC is if you also run a broken OS
1047 that doesn't understand GMT.
1049 config APM_ALLOW_INTS
1050 bool "Allow interrupts during APM BIOS calls"
1053 Normally we disable external interrupts while we are making calls to
1054 the APM BIOS as a measure to lessen the effects of a badly behaving
1055 BIOS implementation. The BIOS should reenable interrupts if it
1056 needs to. Unfortunately, some BIOSes do not -- especially those in
1057 many of the newer IBM Thinkpads. If you experience hangs when you
1058 suspend, try setting this to Y. Otherwise, say N.
1060 config APM_REAL_MODE_POWER_OFF
1061 bool "Use real mode APM BIOS call to power off"
1064 Use real mode APM BIOS calls to switch off the computer. This is
1065 a work-around for a number of buggy BIOSes. Switch this option on if
1066 your computer crashes instead of powering off properly.
1070 source "arch/i386/kernel/cpu/cpufreq/Kconfig"
1075 menu "Bus options (PCI, PCMCIA, EISA, MCA, ISA)"
1077 config X86_VISWS_APIC
1079 depends on X86_VISWS
1082 config X86_LOCAL_APIC
1084 depends on (X86_VISWS || SMP) && !X86_VOYAGER
1089 depends on SMP && !(X86_VISWS || X86_VOYAGER)
1093 bool "PCI support" if !X86_VISWS
1094 depends on !X86_VOYAGER
1095 default y if X86_VISWS
1097 Find out whether you have a PCI motherboard. PCI is the name of a
1098 bus system, i.e. the way the CPU talks to the other stuff inside
1099 your box. Other bus systems are ISA, EISA, MicroChannel (MCA) or
1100 VESA. If you have PCI, say Y, otherwise N.
1102 The PCI-HOWTO, available from
1103 <http://www.tldp.org/docs.html#howto>, contains valuable
1104 information about which PCI hardware does work under Linux and which
1108 prompt "PCI access mode"
1109 depends on PCI && !X86_VISWS
1112 On PCI systems, the BIOS can be used to detect the PCI devices and
1113 determine their configuration. However, some old PCI motherboards
1114 have BIOS bugs and may crash if this is done. Also, some embedded
1115 PCI-based systems don't have any BIOS at all. Linux can also try to
1116 detect the PCI hardware directly without using the BIOS.
1118 With this option, you can specify how Linux should detect the
1119 PCI devices. If you choose "BIOS", the BIOS will be used,
1120 if you choose "Direct", the BIOS won't be used, and if you
1121 choose "MMConfig", then PCI Express MMCONFIG will be used.
1122 If you choose "Any", the kernel will try MMCONFIG, then the
1123 direct access method and falls back to the BIOS if that doesn't
1124 work. If unsure, go with the default, which is "Any".
1129 config PCI_GOMMCONFIG
1142 depends on !X86_VISWS && PCI && (PCI_GOBIOS || PCI_GOANY)
1147 depends on PCI && ((PCI_GODIRECT || PCI_GOANY) || X86_VISWS)
1152 depends on PCI && (PCI_GOMMCONFIG || PCI_GOANY)
1156 source "drivers/pci/Kconfig"
1160 depends on !(X86_VOYAGER || X86_VISWS)
1162 Find out whether you have ISA slots on your motherboard. ISA is the
1163 name of a bus system, i.e. the way the CPU talks to the other stuff
1164 inside your box. Other bus systems are PCI, EISA, MicroChannel
1165 (MCA) or VESA. ISA is an older system, now being displaced by PCI;
1166 newer boards don't support it. If you have ISA, say Y, otherwise N.
1172 The Extended Industry Standard Architecture (EISA) bus was
1173 developed as an open alternative to the IBM MicroChannel bus.
1175 The EISA bus provided some of the features of the IBM MicroChannel
1176 bus while maintaining backward compatibility with cards made for
1177 the older ISA bus. The EISA bus saw limited use between 1988 and
1178 1995 when it was made obsolete by the PCI bus.
1180 Say Y here if you are building a kernel for an EISA-based machine.
1184 source "drivers/eisa/Kconfig"
1188 depends on !(X86_VISWS || X86_VOYAGER)
1190 MicroChannel Architecture is found in some IBM PS/2 machines and
1191 laptops. It is a bus system similar to PCI or ISA. See
1192 <file:Documentation/mca.txt> (and especially the web page given
1193 there) before attempting to build an MCA bus kernel.
1196 depends on X86_VOYAGER
1197 default y if X86_VOYAGER
1199 source "drivers/mca/Kconfig"
1202 tristate "NatSemi SCx200 support"
1203 depends on !X86_VOYAGER
1205 This provides basic support for the National Semiconductor SCx200
1206 processor. Right now this is just a driver for the GPIO pins.
1208 If you don't know what to do here, say N.
1210 This support is also available as a module. If compiled as a
1211 module, it will be called scx200.
1213 source "drivers/pcmcia/Kconfig"
1215 source "drivers/pci/hotplug/Kconfig"
1220 menu "Executable file formats"
1222 source "fs/Kconfig.binfmt"
1226 source "drivers/Kconfig"
1230 source "arch/i386/oprofile/Kconfig"
1233 menu "Kernel hacking"
1236 bool "Kernel debugging"
1238 Say Y here if you are developing drivers or trying to debug and
1239 identify kernel problems.
1242 bool "Early printk" if EMBEDDED
1245 Write kernel log output directly into the VGA buffer or to a serial
1248 This is useful for kernel debugging when your machine crashes very
1249 early before the console code is initialized. For normal operation
1250 it is not recommended because it looks ugly and doesn't cooperate
1251 with klogd/syslogd or the X server. You should normally N here,
1252 unless you want to debug such a crash.
1254 config DEBUG_STACKOVERFLOW
1255 bool "Check for stack overflows"
1256 depends on DEBUG_KERNEL
1258 config DEBUG_STACK_USAGE
1259 bool "Stack utilization instrumentation"
1260 depends on DEBUG_KERNEL
1262 Enables the display of the minimum amount of free stack which each
1263 task has ever had available in the sysrq-T and sysrq-P debug output.
1265 This option will slow down process creation somewhat.
1268 bool "Debug memory allocations"
1269 depends on DEBUG_KERNEL
1271 Say Y here to have the kernel do limited verification on memory
1272 allocation as well as poisoning memory on free to catch use of freed
1276 bool "Magic SysRq key"
1277 depends on DEBUG_KERNEL
1279 If you say Y here, you will have some control over the system even
1280 if the system crashes for example during kernel debugging (e.g., you
1281 will be able to flush the buffer cache to disk, reboot the system
1282 immediately or dump some status information). This is accomplished
1283 by pressing various keys while holding SysRq (Alt+PrintScreen). It
1284 also works on a serial console (on PC hardware at least), if you
1285 send a BREAK and then within 5 seconds a command keypress. The
1286 keys are documented in <file:Documentation/sysrq.txt>. Don't say Y
1287 unless you really know what this hack does.
1289 config DEBUG_SPINLOCK
1290 bool "Spinlock debugging"
1291 depends on DEBUG_KERNEL
1293 Say Y here and build SMP to catch missing spinlock initialization
1294 and certain other kinds of spinlock errors commonly made. This is
1295 best used in conjunction with the NMI watchdog so that spinlock
1296 deadlocks are also debuggable.
1298 config DEBUG_PAGEALLOC
1299 bool "Page alloc debugging"
1300 depends on DEBUG_KERNEL
1302 Unmap pages from the kernel linear mapping after free_pages().
1303 This results in a large slowdown, but helps to find certain types
1304 of memory corruptions.
1306 config DEBUG_HIGHMEM
1307 bool "Highmem debugging"
1308 depends on DEBUG_KERNEL && HIGHMEM
1310 This options enables addition error checking for high memory systems.
1311 Disable for production systems.
1314 bool "Compile the kernel with debug info"
1315 depends on DEBUG_KERNEL
1317 If you say Y here the resulting kernel image will include
1318 debugging info resulting in a larger kernel image.
1319 Say Y here only if you plan to use gdb to debug the kernel.
1320 If you don't debug the kernel, you can say N.
1322 config DEBUG_SPINLOCK_SLEEP
1323 bool "Sleep-inside-spinlock checking"
1325 If you say Y here, various routines which may sleep will become very
1326 noisy if they are called with a spinlock held.
1328 config FRAME_POINTER
1329 bool "Compile the kernel with frame pointers"
1331 If you say Y here the resulting kernel image will be slightly larger
1332 and slower, but it will give very useful debugging information.
1333 If you don't debug the kernel, you can say N, but we may not be able
1334 to solve problems without frame pointers.
1336 config X86_FIND_SMP_CONFIG
1338 depends on X86_LOCAL_APIC || X86_VOYAGER
1343 depends on X86_LOCAL_APIC && !X86_VISWS
1348 source "security/Kconfig"
1350 source "crypto/Kconfig"
1352 source "lib/Kconfig"
1356 depends on SMP && !X86_VOYAGER
1361 depends on SMP && !(X86_VISWS || X86_VOYAGER)
1364 config X86_BIOS_REBOOT
1366 depends on !(X86_VISWS || X86_VOYAGER)
1369 config X86_TRAMPOLINE
1371 depends on X86_SMP || (X86_VOYAGER && SMP)
1374 # std_resources is overridden for pc9800, but that's not
1375 # a currently selectable arch choice
1376 config X86_STD_RESOURCES
1382 depends on X86 && !EMBEDDED