7 option env="KERNELVERSION"
13 default "/lib/modules/$UNAME_RELEASE/.config"
14 default "/etc/kernel-config"
15 default "/boot/config-$UNAME_RELEASE"
16 default "$ARCH_DEFCONFIG"
17 default "arch/$ARCH/defconfig"
26 config BUILDTIME_EXTABLE_SORT
36 depends on BROKEN || !SMP
39 config INIT_ENV_ARG_LIMIT
44 Maximum of each of the number of arguments and environment
45 variables passed to init from the kernel command line.
49 string "Cross-compiler tool prefix"
51 Same as running 'make CROSS_COMPILE=prefix-' but stored for
52 default make runs in this kernel build directory. You don't
53 need to set this unless you want the configured kernel build
54 directory to select the cross-compiler automatically.
57 bool "Compile also drivers which will not load"
60 Some drivers can be compiled on a different platform than they are
61 intended to be run on. Despite they cannot be loaded there (or even
62 when they load they cannot be used due to missing HW support),
63 developers still, opposing to distributors, might want to build such
64 drivers to compile-test them.
66 If you are a developer and want to build everything available, say Y
67 here. If you are a user/distributor, say N here to exclude useless
68 drivers to be distributed.
71 string "Local version - append to kernel release"
73 Append an extra string to the end of your kernel version.
74 This will show up when you type uname, for example.
75 The string you set here will be appended after the contents of
76 any files with a filename matching localversion* in your
77 object and source tree, in that order. Your total string can
78 be a maximum of 64 characters.
80 config LOCALVERSION_AUTO
81 bool "Automatically append version information to the version string"
84 This will try to automatically determine if the current tree is a
85 release tree by looking for git tags that belong to the current
88 A string of the format -gxxxxxxxx will be added to the localversion
89 if a git-based tree is found. The string generated by this will be
90 appended after any matching localversion* files, and after the value
91 set in CONFIG_LOCALVERSION.
93 (The actual string used here is the first eight characters produced
94 by running the command:
96 $ git rev-parse --verify HEAD
98 which is done within the script "scripts/setlocalversion".)
100 config HAVE_KERNEL_GZIP
103 config HAVE_KERNEL_BZIP2
106 config HAVE_KERNEL_LZMA
109 config HAVE_KERNEL_XZ
112 config HAVE_KERNEL_LZO
115 config HAVE_KERNEL_LZ4
119 prompt "Kernel compression mode"
121 depends on HAVE_KERNEL_GZIP || HAVE_KERNEL_BZIP2 || HAVE_KERNEL_LZMA || HAVE_KERNEL_XZ || HAVE_KERNEL_LZO || HAVE_KERNEL_LZ4
123 The linux kernel is a kind of self-extracting executable.
124 Several compression algorithms are available, which differ
125 in efficiency, compression and decompression speed.
126 Compression speed is only relevant when building a kernel.
127 Decompression speed is relevant at each boot.
129 If you have any problems with bzip2 or lzma compressed
130 kernels, mail me (Alain Knaff) <alain@knaff.lu>. (An older
131 version of this functionality (bzip2 only), for 2.4, was
132 supplied by Christian Ludwig)
134 High compression options are mostly useful for users, who
135 are low on disk space (embedded systems), but for whom ram
138 If in doubt, select 'gzip'
142 depends on HAVE_KERNEL_GZIP
144 The old and tried gzip compression. It provides a good balance
145 between compression ratio and decompression speed.
149 depends on HAVE_KERNEL_BZIP2
151 Its compression ratio and speed is intermediate.
152 Decompression speed is slowest among the choices. The kernel
153 size is about 10% smaller with bzip2, in comparison to gzip.
154 Bzip2 uses a large amount of memory. For modern kernels you
155 will need at least 8MB RAM or more for booting.
159 depends on HAVE_KERNEL_LZMA
161 This compression algorithm's ratio is best. Decompression speed
162 is between gzip and bzip2. Compression is slowest.
163 The kernel size is about 33% smaller with LZMA in comparison to gzip.
167 depends on HAVE_KERNEL_XZ
169 XZ uses the LZMA2 algorithm and instruction set specific
170 BCJ filters which can improve compression ratio of executable
171 code. The size of the kernel is about 30% smaller with XZ in
172 comparison to gzip. On architectures for which there is a BCJ
173 filter (i386, x86_64, ARM, IA-64, PowerPC, and SPARC), XZ
174 will create a few percent smaller kernel than plain LZMA.
176 The speed is about the same as with LZMA: The decompression
177 speed of XZ is better than that of bzip2 but worse than gzip
178 and LZO. Compression is slow.
182 depends on HAVE_KERNEL_LZO
184 Its compression ratio is the poorest among the choices. The kernel
185 size is about 10% bigger than gzip; however its speed
186 (both compression and decompression) is the fastest.
190 depends on HAVE_KERNEL_LZ4
192 LZ4 is an LZ77-type compressor with a fixed, byte-oriented encoding.
193 A preliminary version of LZ4 de/compression tool is available at
194 <https://code.google.com/p/lz4/>.
196 Its compression ratio is worse than LZO. The size of the kernel
197 is about 8% bigger than LZO. But the decompression speed is
202 config DEFAULT_HOSTNAME
203 string "Default hostname"
206 This option determines the default system hostname before userspace
207 calls sethostname(2). The kernel traditionally uses "(none)" here,
208 but you may wish to use a different default here to make a minimal
209 system more usable with less configuration.
212 bool "Support for paging of anonymous memory (swap)"
213 depends on MMU && BLOCK
216 This option allows you to choose whether you want to have support
217 for so called swap devices or swap files in your kernel that are
218 used to provide more virtual memory than the actual RAM present
219 in your computer. If unsure say Y.
224 Inter Process Communication is a suite of library functions and
225 system calls which let processes (running programs) synchronize and
226 exchange information. It is generally considered to be a good thing,
227 and some programs won't run unless you say Y here. In particular, if
228 you want to run the DOS emulator dosemu under Linux (read the
229 DOSEMU-HOWTO, available from <http://www.tldp.org/docs.html#howto>),
230 you'll need to say Y here.
232 You can find documentation about IPC with "info ipc" and also in
233 section 6.4 of the Linux Programmer's Guide, available from
234 <http://www.tldp.org/guides.html>.
236 config SYSVIPC_SYSCTL
243 bool "POSIX Message Queues"
246 POSIX variant of message queues is a part of IPC. In POSIX message
247 queues every message has a priority which decides about succession
248 of receiving it by a process. If you want to compile and run
249 programs written e.g. for Solaris with use of its POSIX message
250 queues (functions mq_*) say Y here.
252 POSIX message queues are visible as a filesystem called 'mqueue'
253 and can be mounted somewhere if you want to do filesystem
254 operations on message queues.
258 config POSIX_MQUEUE_SYSCTL
260 depends on POSIX_MQUEUE
265 bool "open by fhandle syscalls"
268 If you say Y here, a user level program will be able to map
269 file names to handle and then later use the handle for
270 different file system operations. This is useful in implementing
271 userspace file servers, which now track files using handles instead
272 of names. The handle would remain the same even if file names
273 get renamed. Enables open_by_handle_at(2) and name_to_handle_at(2)
277 bool "Auditing support"
280 Enable auditing infrastructure that can be used with another
281 kernel subsystem, such as SELinux (which requires this for
282 logging of avc messages output). Does not do system-call
283 auditing without CONFIG_AUDITSYSCALL.
286 bool "Enable system-call auditing support"
287 depends on AUDIT && (X86 || PPC || S390 || IA64 || UML || SPARC64 || SUPERH || (ARM && AEABI && !OABI_COMPAT))
288 default y if SECURITY_SELINUX
290 Enable low-overhead system-call auditing infrastructure that
291 can be used independently or with another kernel subsystem,
296 depends on AUDITSYSCALL
301 depends on AUDITSYSCALL
304 source "kernel/irq/Kconfig"
305 source "kernel/time/Kconfig"
307 menu "CPU/Task time and stats accounting"
309 config VIRT_CPU_ACCOUNTING
313 prompt "Cputime accounting"
314 default TICK_CPU_ACCOUNTING if !PPC64
315 default VIRT_CPU_ACCOUNTING_NATIVE if PPC64
317 # Kind of a stub config for the pure tick based cputime accounting
318 config TICK_CPU_ACCOUNTING
319 bool "Simple tick based cputime accounting"
320 depends on !S390 && !NO_HZ_FULL
322 This is the basic tick based cputime accounting that maintains
323 statistics about user, system and idle time spent on per jiffies
328 config VIRT_CPU_ACCOUNTING_NATIVE
329 bool "Deterministic task and CPU time accounting"
330 depends on HAVE_VIRT_CPU_ACCOUNTING && !NO_HZ_FULL
331 select VIRT_CPU_ACCOUNTING
333 Select this option to enable more accurate task and CPU time
334 accounting. This is done by reading a CPU counter on each
335 kernel entry and exit and on transitions within the kernel
336 between system, softirq and hardirq state, so there is a
337 small performance impact. In the case of s390 or IBM POWER > 5,
338 this also enables accounting of stolen time on logically-partitioned
341 config VIRT_CPU_ACCOUNTING_GEN
342 bool "Full dynticks CPU time accounting"
343 depends on HAVE_CONTEXT_TRACKING && 64BIT
344 select VIRT_CPU_ACCOUNTING
345 select CONTEXT_TRACKING
347 Select this option to enable task and CPU time accounting on full
348 dynticks systems. This accounting is implemented by watching every
349 kernel-user boundaries using the context tracking subsystem.
350 The accounting is thus performed at the expense of some significant
353 For now this is only useful if you are working on the full
354 dynticks subsystem development.
358 config IRQ_TIME_ACCOUNTING
359 bool "Fine granularity task level IRQ time accounting"
360 depends on HAVE_IRQ_TIME_ACCOUNTING && !NO_HZ_FULL
362 Select this option to enable fine granularity task irq time
363 accounting. This is done by reading a timestamp on each
364 transitions between softirq and hardirq state, so there can be a
365 small performance impact.
367 If in doubt, say N here.
371 config BSD_PROCESS_ACCT
372 bool "BSD Process Accounting"
374 If you say Y here, a user level program will be able to instruct the
375 kernel (via a special system call) to write process accounting
376 information to a file: whenever a process exits, information about
377 that process will be appended to the file by the kernel. The
378 information includes things such as creation time, owning user,
379 command name, memory usage, controlling terminal etc. (the complete
380 list is in the struct acct in <file:include/linux/acct.h>). It is
381 up to the user level program to do useful things with this
382 information. This is generally a good idea, so say Y.
384 config BSD_PROCESS_ACCT_V3
385 bool "BSD Process Accounting version 3 file format"
386 depends on BSD_PROCESS_ACCT
389 If you say Y here, the process accounting information is written
390 in a new file format that also logs the process IDs of each
391 process and it's parent. Note that this file format is incompatible
392 with previous v0/v1/v2 file formats, so you will need updated tools
393 for processing it. A preliminary version of these tools is available
394 at <http://www.gnu.org/software/acct/>.
397 bool "Export task/process statistics through netlink"
401 Export selected statistics for tasks/processes through the
402 generic netlink interface. Unlike BSD process accounting, the
403 statistics are available during the lifetime of tasks/processes as
404 responses to commands. Like BSD accounting, they are sent to user
409 config TASK_DELAY_ACCT
410 bool "Enable per-task delay accounting"
413 Collect information on time spent by a task waiting for system
414 resources like cpu, synchronous block I/O completion and swapping
415 in pages. Such statistics can help in setting a task's priorities
416 relative to other tasks for cpu, io, rss limits etc.
421 bool "Enable extended accounting over taskstats"
424 Collect extended task accounting data and send the data
425 to userland for processing over the taskstats interface.
429 config TASK_IO_ACCOUNTING
430 bool "Enable per-task storage I/O accounting"
431 depends on TASK_XACCT
433 Collect information on the number of bytes of storage I/O which this
438 endmenu # "CPU/Task time and stats accounting"
443 prompt "RCU Implementation"
447 bool "Tree-based hierarchical RCU"
448 depends on !PREEMPT && SMP
451 This option selects the RCU implementation that is
452 designed for very large SMP system with hundreds or
453 thousands of CPUs. It also scales down nicely to
456 config TREE_PREEMPT_RCU
457 bool "Preemptible tree-based hierarchical RCU"
461 This option selects the RCU implementation that is
462 designed for very large SMP systems with hundreds or
463 thousands of CPUs, but for which real-time response
464 is also required. It also scales down nicely to
467 Select this option if you are unsure.
470 bool "UP-only small-memory-footprint RCU"
471 depends on !PREEMPT && !SMP
473 This option selects the RCU implementation that is
474 designed for UP systems from which real-time response
475 is not required. This option greatly reduces the
476 memory footprint of RCU.
481 def_bool TREE_PREEMPT_RCU
483 This option enables preemptible-RCU code that is common between
484 the TREE_PREEMPT_RCU and TINY_PREEMPT_RCU implementations.
486 config RCU_STALL_COMMON
487 def_bool ( TREE_RCU || TREE_PREEMPT_RCU || RCU_TRACE )
489 This option enables RCU CPU stall code that is common between
490 the TINY and TREE variants of RCU. The purpose is to allow
491 the tiny variants to disable RCU CPU stall warnings, while
492 making these warnings mandatory for the tree variants.
494 config CONTEXT_TRACKING
498 bool "Consider userspace as in RCU extended quiescent state"
499 depends on HAVE_CONTEXT_TRACKING && SMP
500 select CONTEXT_TRACKING
502 This option sets hooks on kernel / userspace boundaries and
503 puts RCU in extended quiescent state when the CPU runs in
504 userspace. It means that when a CPU runs in userspace, it is
505 excluded from the global RCU state machine and thus doesn't
506 try to keep the timer tick on for RCU.
508 Unless you want to hack and help the development of the full
509 dynticks mode, you shouldn't enable this option. It also
510 adds unnecessary overhead.
514 config CONTEXT_TRACKING_FORCE
515 bool "Force context tracking"
516 depends on CONTEXT_TRACKING
517 default y if !NO_HZ_FULL
519 The major pre-requirement for full dynticks to work is to
520 support the context tracking subsystem. But there are also
521 other dependencies to provide in order to make the full
524 This option stands for testing when an arch implements the
525 context tracking backend but doesn't yet fullfill all the
526 requirements to make the full dynticks feature working.
527 Without the full dynticks, there is no way to test the support
528 for context tracking and the subsystems that rely on it: RCU
529 userspace extended quiescent state and tickless cputime
530 accounting. This option copes with the absence of the full
531 dynticks subsystem by forcing the context tracking on all
534 Say Y only if you're working on the developpement of an
535 architecture backend for the context tracking.
537 Say N otherwise, this option brings an overhead that you
538 don't want in production.
542 int "Tree-based hierarchical RCU fanout value"
545 depends on TREE_RCU || TREE_PREEMPT_RCU
549 This option controls the fanout of hierarchical implementations
550 of RCU, allowing RCU to work efficiently on machines with
551 large numbers of CPUs. This value must be at least the fourth
552 root of NR_CPUS, which allows NR_CPUS to be insanely large.
553 The default value of RCU_FANOUT should be used for production
554 systems, but if you are stress-testing the RCU implementation
555 itself, small RCU_FANOUT values allow you to test large-system
556 code paths on small(er) systems.
558 Select a specific number if testing RCU itself.
559 Take the default if unsure.
561 config RCU_FANOUT_LEAF
562 int "Tree-based hierarchical RCU leaf-level fanout value"
563 range 2 RCU_FANOUT if 64BIT
564 range 2 RCU_FANOUT if !64BIT
565 depends on TREE_RCU || TREE_PREEMPT_RCU
568 This option controls the leaf-level fanout of hierarchical
569 implementations of RCU, and allows trading off cache misses
570 against lock contention. Systems that synchronize their
571 scheduling-clock interrupts for energy-efficiency reasons will
572 want the default because the smaller leaf-level fanout keeps
573 lock contention levels acceptably low. Very large systems
574 (hundreds or thousands of CPUs) will instead want to set this
575 value to the maximum value possible in order to reduce the
576 number of cache misses incurred during RCU's grace-period
577 initialization. These systems tend to run CPU-bound, and thus
578 are not helped by synchronized interrupts, and thus tend to
579 skew them, which reduces lock contention enough that large
580 leaf-level fanouts work well.
582 Select a specific number if testing RCU itself.
584 Select the maximum permissible value for large systems.
586 Take the default if unsure.
588 config RCU_FANOUT_EXACT
589 bool "Disable tree-based hierarchical RCU auto-balancing"
590 depends on TREE_RCU || TREE_PREEMPT_RCU
593 This option forces use of the exact RCU_FANOUT value specified,
594 regardless of imbalances in the hierarchy. This is useful for
595 testing RCU itself, and might one day be useful on systems with
596 strong NUMA behavior.
598 Without RCU_FANOUT_EXACT, the code will balance the hierarchy.
602 config RCU_FAST_NO_HZ
603 bool "Accelerate last non-dyntick-idle CPU's grace periods"
604 depends on NO_HZ_COMMON && SMP
607 This option permits CPUs to enter dynticks-idle state even if
608 they have RCU callbacks queued, and prevents RCU from waking
609 these CPUs up more than roughly once every four jiffies (by
610 default, you can adjust this using the rcutree.rcu_idle_gp_delay
611 parameter), thus improving energy efficiency. On the other
612 hand, this option increases the duration of RCU grace periods,
613 for example, slowing down synchronize_rcu().
615 Say Y if energy efficiency is critically important, and you
616 don't care about increased grace-period durations.
618 Say N if you are unsure.
620 config TREE_RCU_TRACE
621 def_bool RCU_TRACE && ( TREE_RCU || TREE_PREEMPT_RCU )
624 This option provides tracing for the TREE_RCU and
625 TREE_PREEMPT_RCU implementations, permitting Makefile to
626 trivially select kernel/rcutree_trace.c.
629 bool "Enable RCU priority boosting"
630 depends on RT_MUTEXES && PREEMPT_RCU
633 This option boosts the priority of preempted RCU readers that
634 block the current preemptible RCU grace period for too long.
635 This option also prevents heavy loads from blocking RCU
636 callback invocation for all flavors of RCU.
638 Say Y here if you are working with real-time apps or heavy loads
639 Say N here if you are unsure.
641 config RCU_BOOST_PRIO
642 int "Real-time priority to boost RCU readers to"
647 This option specifies the real-time priority to which long-term
648 preempted RCU readers are to be boosted. If you are working
649 with a real-time application that has one or more CPU-bound
650 threads running at a real-time priority level, you should set
651 RCU_BOOST_PRIO to a priority higher then the highest-priority
652 real-time CPU-bound thread. The default RCU_BOOST_PRIO value
653 of 1 is appropriate in the common case, which is real-time
654 applications that do not have any CPU-bound threads.
656 Some real-time applications might not have a single real-time
657 thread that saturates a given CPU, but instead might have
658 multiple real-time threads that, taken together, fully utilize
659 that CPU. In this case, you should set RCU_BOOST_PRIO to
660 a priority higher than the lowest-priority thread that is
661 conspiring to prevent the CPU from running any non-real-time
662 tasks. For example, if one thread at priority 10 and another
663 thread at priority 5 are between themselves fully consuming
664 the CPU time on a given CPU, then RCU_BOOST_PRIO should be
665 set to priority 6 or higher.
667 Specify the real-time priority, or take the default if unsure.
669 config RCU_BOOST_DELAY
670 int "Milliseconds to delay boosting after RCU grace-period start"
675 This option specifies the time to wait after the beginning of
676 a given grace period before priority-boosting preempted RCU
677 readers blocking that grace period. Note that any RCU reader
678 blocking an expedited RCU grace period is boosted immediately.
680 Accept the default if unsure.
683 bool "Offload RCU callback processing from boot-selected CPUs"
684 depends on TREE_RCU || TREE_PREEMPT_RCU
687 Use this option to reduce OS jitter for aggressive HPC or
688 real-time workloads. It can also be used to offload RCU
689 callback invocation to energy-efficient CPUs in battery-powered
690 asymmetric multiprocessors.
692 This option offloads callback invocation from the set of
693 CPUs specified at boot time by the rcu_nocbs parameter.
694 For each such CPU, a kthread ("rcuox/N") will be created to
695 invoke callbacks, where the "N" is the CPU being offloaded,
696 and where the "x" is "b" for RCU-bh, "p" for RCU-preempt, and
697 "s" for RCU-sched. Nothing prevents this kthread from running
698 on the specified CPUs, but (1) the kthreads may be preempted
699 between each callback, and (2) affinity or cgroups can be used
700 to force the kthreads to run on whatever set of CPUs is desired.
702 Say Y here if you want to help to debug reduced OS jitter.
703 Say N here if you are unsure.
706 prompt "Build-forced no-CBs CPUs"
707 default RCU_NOCB_CPU_NONE
709 This option allows no-CBs CPUs (whose RCU callbacks are invoked
710 from kthreads rather than from softirq context) to be specified
711 at build time. Additional no-CBs CPUs may be specified by
712 the rcu_nocbs= boot parameter.
714 config RCU_NOCB_CPU_NONE
715 bool "No build_forced no-CBs CPUs"
716 depends on RCU_NOCB_CPU && !NO_HZ_FULL
718 This option does not force any of the CPUs to be no-CBs CPUs.
719 Only CPUs designated by the rcu_nocbs= boot parameter will be
720 no-CBs CPUs, whose RCU callbacks will be invoked by per-CPU
721 kthreads whose names begin with "rcuo". All other CPUs will
722 invoke their own RCU callbacks in softirq context.
724 Select this option if you want to choose no-CBs CPUs at
725 boot time, for example, to allow testing of different no-CBs
726 configurations without having to rebuild the kernel each time.
728 config RCU_NOCB_CPU_ZERO
729 bool "CPU 0 is a build_forced no-CBs CPU"
730 depends on RCU_NOCB_CPU && !NO_HZ_FULL
732 This option forces CPU 0 to be a no-CBs CPU, so that its RCU
733 callbacks are invoked by a per-CPU kthread whose name begins
734 with "rcuo". Additional CPUs may be designated as no-CBs
735 CPUs using the rcu_nocbs= boot parameter will be no-CBs CPUs.
736 All other CPUs will invoke their own RCU callbacks in softirq
739 Select this if CPU 0 needs to be a no-CBs CPU for real-time
740 or energy-efficiency reasons, but the real reason it exists
741 is to ensure that randconfig testing covers mixed systems.
743 config RCU_NOCB_CPU_ALL
744 bool "All CPUs are build_forced no-CBs CPUs"
745 depends on RCU_NOCB_CPU
747 This option forces all CPUs to be no-CBs CPUs. The rcu_nocbs=
748 boot parameter will be ignored. All CPUs' RCU callbacks will
749 be executed in the context of per-CPU rcuo kthreads created for
750 this purpose. Assuming that the kthreads whose names start with
751 "rcuo" are bound to "housekeeping" CPUs, this reduces OS jitter
752 on the remaining CPUs, but might decrease memory locality during
753 RCU-callback invocation, thus potentially degrading throughput.
755 Select this if all CPUs need to be no-CBs CPUs for real-time
756 or energy-efficiency reasons.
760 endmenu # "RCU Subsystem"
763 tristate "Kernel .config support"
765 This option enables the complete Linux kernel ".config" file
766 contents to be saved in the kernel. It provides documentation
767 of which kernel options are used in a running kernel or in an
768 on-disk kernel. This information can be extracted from the kernel
769 image file with the script scripts/extract-ikconfig and used as
770 input to rebuild the current kernel or to build another kernel.
771 It can also be extracted from a running kernel by reading
772 /proc/config.gz if enabled (below).
775 bool "Enable access to .config through /proc/config.gz"
776 depends on IKCONFIG && PROC_FS
778 This option enables access to the kernel configuration file
779 through /proc/config.gz.
782 int "Kernel log buffer size (16 => 64KB, 17 => 128KB)"
786 Select kernel log buffer size as a power of 2.
796 # Architectures with an unreliable sched_clock() should select this:
798 config HAVE_UNSTABLE_SCHED_CLOCK
801 config GENERIC_SCHED_CLOCK
805 # For architectures that want to enable the support for NUMA-affine scheduler
808 config ARCH_SUPPORTS_NUMA_BALANCING
811 # For architectures that (ab)use NUMA to represent different memory regions
812 # all cpu-local but of different latencies, such as SuperH.
814 config ARCH_WANT_NUMA_VARIABLE_LOCALITY
818 # For architectures that are willing to define _PAGE_NUMA as _PAGE_PROTNONE
819 config ARCH_WANTS_PROT_NUMA_PROT_NONE
822 config ARCH_USES_NUMA_PROT_NONE
825 depends on ARCH_WANTS_PROT_NUMA_PROT_NONE
826 depends on NUMA_BALANCING
828 config NUMA_BALANCING_DEFAULT_ENABLED
829 bool "Automatically enable NUMA aware memory/task placement"
831 depends on NUMA_BALANCING
833 If set, autonumic NUMA balancing will be enabled if running on a NUMA
836 config NUMA_BALANCING
837 bool "Memory placement aware NUMA scheduler"
838 depends on ARCH_SUPPORTS_NUMA_BALANCING
839 depends on !ARCH_WANT_NUMA_VARIABLE_LOCALITY
840 depends on SMP && NUMA && MIGRATION
842 This option adds support for automatic NUMA aware memory/task placement.
843 The mechanism is quite primitive and is based on migrating memory when
844 it is references to the node the task is running on.
846 This system will be inactive on UMA systems.
849 boolean "Control Group support"
852 This option adds support for grouping sets of processes together, for
853 use with process control subsystems such as Cpusets, CFS, memory
854 controls or device isolation.
856 - Documentation/scheduler/sched-design-CFS.txt (CFS)
857 - Documentation/cgroups/ (features for grouping, isolation
858 and resource control)
865 bool "Example debug cgroup subsystem"
868 This option enables a simple cgroup subsystem that
869 exports useful debugging information about the cgroups
874 config CGROUP_FREEZER
875 bool "Freezer cgroup subsystem"
877 Provides a way to freeze and unfreeze all tasks in a
881 bool "Device controller for cgroups"
883 Provides a cgroup implementing whitelists for devices which
884 a process in the cgroup can mknod or open.
887 bool "Cpuset support"
889 This option will let you create and manage CPUSETs which
890 allow dynamically partitioning a system into sets of CPUs and
891 Memory Nodes and assigning tasks to run only within those sets.
892 This is primarily useful on large SMP or NUMA systems.
896 config PROC_PID_CPUSET
897 bool "Include legacy /proc/<pid>/cpuset file"
901 config CGROUP_CPUACCT
902 bool "Simple CPU accounting cgroup subsystem"
904 Provides a simple Resource Controller for monitoring the
905 total CPU consumed by the tasks in a cgroup.
907 config RESOURCE_COUNTERS
908 bool "Resource counters"
910 This option enables controller independent resource accounting
911 infrastructure that works with cgroups.
914 bool "Memory Resource Controller for Control Groups"
915 depends on RESOURCE_COUNTERS
918 Provides a memory resource controller that manages both anonymous
919 memory and page cache. (See Documentation/cgroups/memory.txt)
921 Note that setting this option increases fixed memory overhead
922 associated with each page of memory in the system. By this,
923 8(16)bytes/PAGE_SIZE on 32(64)bit system will be occupied by memory
924 usage tracking struct at boot. Total amount of this is printed out
927 Only enable when you're ok with these trade offs and really
928 sure you need the memory resource controller. Even when you enable
929 this, you can set "cgroup_disable=memory" at your boot option to
930 disable memory resource controller and you can avoid overheads.
931 (and lose benefits of memory resource controller)
933 This config option also selects MM_OWNER config option, which
934 could in turn add some fork/exit overhead.
937 bool "Memory Resource Controller Swap Extension"
938 depends on MEMCG && SWAP
940 Add swap management feature to memory resource controller. When you
941 enable this, you can limit mem+swap usage per cgroup. In other words,
942 when you disable this, memory resource controller has no cares to
943 usage of swap...a process can exhaust all of the swap. This extension
944 is useful when you want to avoid exhaustion swap but this itself
945 adds more overheads and consumes memory for remembering information.
946 Especially if you use 32bit system or small memory system, please
947 be careful about enabling this. When memory resource controller
948 is disabled by boot option, this will be automatically disabled and
949 there will be no overhead from this. Even when you set this config=y,
950 if boot option "swapaccount=0" is set, swap will not be accounted.
951 Now, memory usage of swap_cgroup is 2 bytes per entry. If swap page
952 size is 4096bytes, 512k per 1Gbytes of swap.
953 config MEMCG_SWAP_ENABLED
954 bool "Memory Resource Controller Swap Extension enabled by default"
955 depends on MEMCG_SWAP
958 Memory Resource Controller Swap Extension comes with its price in
959 a bigger memory consumption. General purpose distribution kernels
960 which want to enable the feature but keep it disabled by default
961 and let the user enable it by swapaccount=1 boot command line
962 parameter should have this option unselected.
963 For those who want to have the feature enabled by default should
964 select this option (if, for some reason, they need to disable it
965 then swapaccount=0 does the trick).
967 bool "Memory Resource Controller Kernel Memory accounting"
969 depends on SLUB || SLAB
971 The Kernel Memory extension for Memory Resource Controller can limit
972 the amount of memory used by kernel objects in the system. Those are
973 fundamentally different from the entities handled by the standard
974 Memory Controller, which are page-based, and can be swapped. Users of
975 the kmem extension can use it to guarantee that no group of processes
976 will ever exhaust kernel resources alone.
978 config CGROUP_HUGETLB
979 bool "HugeTLB Resource Controller for Control Groups"
980 depends on RESOURCE_COUNTERS && HUGETLB_PAGE
983 Provides a cgroup Resource Controller for HugeTLB pages.
984 When you enable this, you can put a per cgroup limit on HugeTLB usage.
985 The limit is enforced during page fault. Since HugeTLB doesn't
986 support page reclaim, enforcing the limit at page fault time implies
987 that, the application will get SIGBUS signal if it tries to access
988 HugeTLB pages beyond its limit. This requires the application to know
989 beforehand how much HugeTLB pages it would require for its use. The
990 control group is tracked in the third page lru pointer. This means
991 that we cannot use the controller with huge page less than 3 pages.
994 bool "Enable perf_event per-cpu per-container group (cgroup) monitoring"
995 depends on PERF_EVENTS && CGROUPS
997 This option extends the per-cpu mode to restrict monitoring to
998 threads which belong to the cgroup specified and run on the
1003 menuconfig CGROUP_SCHED
1004 bool "Group CPU scheduler"
1007 This feature lets CPU scheduler recognize task groups and control CPU
1008 bandwidth allocation to such task groups. It uses cgroups to group
1012 config FAIR_GROUP_SCHED
1013 bool "Group scheduling for SCHED_OTHER"
1014 depends on CGROUP_SCHED
1015 default CGROUP_SCHED
1017 config CFS_BANDWIDTH
1018 bool "CPU bandwidth provisioning for FAIR_GROUP_SCHED"
1019 depends on FAIR_GROUP_SCHED
1022 This option allows users to define CPU bandwidth rates (limits) for
1023 tasks running within the fair group scheduler. Groups with no limit
1024 set are considered to be unconstrained and will run with no
1026 See tip/Documentation/scheduler/sched-bwc.txt for more information.
1028 config RT_GROUP_SCHED
1029 bool "Group scheduling for SCHED_RR/FIFO"
1030 depends on CGROUP_SCHED
1033 This feature lets you explicitly allocate real CPU bandwidth
1034 to task groups. If enabled, it will also make it impossible to
1035 schedule realtime tasks for non-root users until you allocate
1036 realtime bandwidth for them.
1037 See Documentation/scheduler/sched-rt-group.txt for more information.
1042 bool "Block IO controller"
1046 Generic block IO controller cgroup interface. This is the common
1047 cgroup interface which should be used by various IO controlling
1050 Currently, CFQ IO scheduler uses it to recognize task groups and
1051 control disk bandwidth allocation (proportional time slice allocation)
1052 to such task groups. It is also used by bio throttling logic in
1053 block layer to implement upper limit in IO rates on a device.
1055 This option only enables generic Block IO controller infrastructure.
1056 One needs to also enable actual IO controlling logic/policy. For
1057 enabling proportional weight division of disk bandwidth in CFQ, set
1058 CONFIG_CFQ_GROUP_IOSCHED=y; for enabling throttling policy, set
1059 CONFIG_BLK_DEV_THROTTLING=y.
1061 See Documentation/cgroups/blkio-controller.txt for more information.
1063 config DEBUG_BLK_CGROUP
1064 bool "Enable Block IO controller debugging"
1065 depends on BLK_CGROUP
1068 Enable some debugging help. Currently it exports additional stat
1069 files in a cgroup which can be useful for debugging.
1073 config CHECKPOINT_RESTORE
1074 bool "Checkpoint/restore support" if EXPERT
1077 Enables additional kernel features in a sake of checkpoint/restore.
1078 In particular it adds auxiliary prctl codes to setup process text,
1079 data and heap segment sizes, and a few additional /proc filesystem
1082 If unsure, say N here.
1084 menuconfig NAMESPACES
1085 bool "Namespaces support" if EXPERT
1088 Provides the way to make tasks work with different objects using
1089 the same id. For example same IPC id may refer to different objects
1090 or same user id or pid may refer to different tasks when used in
1091 different namespaces.
1096 bool "UTS namespace"
1099 In this namespace tasks see different info provided with the
1103 bool "IPC namespace"
1104 depends on (SYSVIPC || POSIX_MQUEUE)
1107 In this namespace tasks work with IPC ids which correspond to
1108 different IPC objects in different namespaces.
1111 bool "User namespace"
1112 select UIDGID_STRICT_TYPE_CHECKS
1116 This allows containers, i.e. vservers, to use user namespaces
1117 to provide different user info for different servers.
1119 When user namespaces are enabled in the kernel it is
1120 recommended that the MEMCG and MEMCG_KMEM options also be
1121 enabled and that user-space use the memory control groups to
1122 limit the amount of memory a memory unprivileged users can
1128 bool "PID Namespaces"
1131 Support process id namespaces. This allows having multiple
1132 processes with the same pid as long as they are in different
1133 pid namespaces. This is a building block of containers.
1136 bool "Network namespace"
1140 Allow user space to create what appear to be multiple instances
1141 of the network stack.
1145 config UIDGID_STRICT_TYPE_CHECKS
1146 bool "Require conversions between uid/gids and their internal representation"
1149 While the nececessary conversions are being added to all subsystems this option allows
1150 the code to continue to build for unconverted subsystems.
1152 Say Y here if you want the strict type checking enabled
1154 config SCHED_AUTOGROUP
1155 bool "Automatic process group scheduling"
1159 select FAIR_GROUP_SCHED
1161 This option optimizes the scheduler for common desktop workloads by
1162 automatically creating and populating task groups. This separation
1163 of workloads isolates aggressive CPU burners (like build jobs) from
1164 desktop applications. Task group autogeneration is currently based
1170 config SYSFS_DEPRECATED
1171 bool "Enable deprecated sysfs features to support old userspace tools"
1175 This option adds code that switches the layout of the "block" class
1176 devices, to not show up in /sys/class/block/, but only in
1179 This switch is only active when the sysfs.deprecated=1 boot option is
1180 passed or the SYSFS_DEPRECATED_V2 option is set.
1182 This option allows new kernels to run on old distributions and tools,
1183 which might get confused by /sys/class/block/. Since 2007/2008 all
1184 major distributions and tools handle this just fine.
1186 Recent distributions and userspace tools after 2009/2010 depend on
1187 the existence of /sys/class/block/, and will not work with this
1190 Only if you are using a new kernel on an old distribution, you might
1193 config SYSFS_DEPRECATED_V2
1194 bool "Enable deprecated sysfs features by default"
1197 depends on SYSFS_DEPRECATED
1199 Enable deprecated sysfs by default.
1201 See the CONFIG_SYSFS_DEPRECATED option for more details about this
1204 Only if you are using a new kernel on an old distribution, you might
1205 need to say Y here. Even then, odds are you would not need it
1206 enabled, you can always pass the boot option if absolutely necessary.
1209 bool "Kernel->user space relay support (formerly relayfs)"
1211 This option enables support for relay interface support in
1212 certain file systems (such as debugfs).
1213 It is designed to provide an efficient mechanism for tools and
1214 facilities to relay large amounts of data from kernel space to
1219 config BLK_DEV_INITRD
1220 bool "Initial RAM filesystem and RAM disk (initramfs/initrd) support"
1221 depends on BROKEN || !FRV
1223 The initial RAM filesystem is a ramfs which is loaded by the
1224 boot loader (loadlin or lilo) and that is mounted as root
1225 before the normal boot procedure. It is typically used to
1226 load modules needed to mount the "real" root file system,
1227 etc. See <file:Documentation/initrd.txt> for details.
1229 If RAM disk support (BLK_DEV_RAM) is also included, this
1230 also enables initial RAM disk (initrd) support and adds
1231 15 Kbytes (more on some other architectures) to the kernel size.
1237 source "usr/Kconfig"
1241 config CC_OPTIMIZE_FOR_SIZE
1242 bool "Optimize for size"
1244 Enabling this option will pass "-Os" instead of "-O2" to gcc
1245 resulting in a smaller kernel.
1258 config SYSCTL_EXCEPTION_TRACE
1261 Enable support for /proc/sys/debug/exception-trace.
1263 config SYSCTL_ARCH_UNALIGN_NO_WARN
1266 Enable support for /proc/sys/kernel/ignore-unaligned-usertrap
1267 Allows arch to define/use @no_unaligned_warning to possibly warn
1268 about unaligned access emulation going on under the hood.
1270 config SYSCTL_ARCH_UNALIGN_ALLOW
1273 Enable support for /proc/sys/kernel/unaligned-trap
1274 Allows arches to define/use @unaligned_enabled to runtime toggle
1275 the unaligned access emulation.
1276 see arch/parisc/kernel/unaligned.c for reference
1278 config HAVE_PCSPKR_PLATFORM
1282 bool "Configure standard kernel features (expert users)"
1283 # Unhide debug options, to make the on-by-default options visible
1286 This option allows certain base kernel options and settings
1287 to be disabled or tweaked. This is for specialized
1288 environments which can tolerate a "non-standard" kernel.
1289 Only use this if you really know what you are doing.
1292 bool "Enable 16-bit UID system calls" if EXPERT
1293 depends on HAVE_UID16
1296 This enables the legacy 16-bit UID syscall wrappers.
1298 config SYSCTL_SYSCALL
1299 bool "Sysctl syscall support" if EXPERT
1300 depends on PROC_SYSCTL
1304 sys_sysctl uses binary paths that have been found challenging
1305 to properly maintain and use. The interface in /proc/sys
1306 using paths with ascii names is now the primary path to this
1309 Almost nothing using the binary sysctl interface so if you are
1310 trying to save some space it is probably safe to disable this,
1311 making your kernel marginally smaller.
1313 If unsure say N here.
1316 bool "Load all symbols for debugging/ksymoops" if EXPERT
1319 Say Y here to let the kernel print out symbolic crash information and
1320 symbolic stack backtraces. This increases the size of the kernel
1321 somewhat, as all symbols have to be loaded into the kernel image.
1324 bool "Include all symbols in kallsyms"
1325 depends on DEBUG_KERNEL && KALLSYMS
1327 Normally kallsyms only contains the symbols of functions for nicer
1328 OOPS messages and backtraces (i.e., symbols from the text and inittext
1329 sections). This is sufficient for most cases. And only in very rare
1330 cases (e.g., when a debugger is used) all symbols are required (e.g.,
1331 names of variables from the data sections, etc).
1333 This option makes sure that all symbols are loaded into the kernel
1334 image (i.e., symbols from all sections) in cost of increased kernel
1335 size (depending on the kernel configuration, it may be 300KiB or
1336 something like this).
1338 Say N unless you really need all symbols.
1342 bool "Enable support for printk" if EXPERT
1345 This option enables normal printk support. Removing it
1346 eliminates most of the message strings from the kernel image
1347 and makes the kernel more or less silent. As this makes it
1348 very difficult to diagnose system problems, saying N here is
1349 strongly discouraged.
1352 bool "BUG() support" if EXPERT
1355 Disabling this option eliminates support for BUG and WARN, reducing
1356 the size of your kernel image and potentially quietly ignoring
1357 numerous fatal conditions. You should only consider disabling this
1358 option for embedded systems with no facilities for reporting errors.
1364 bool "Enable ELF core dumps" if EXPERT
1366 Enable support for generating core dumps. Disabling saves about 4k.
1369 config PCSPKR_PLATFORM
1370 bool "Enable PC-Speaker support" if EXPERT
1371 depends on HAVE_PCSPKR_PLATFORM
1375 This option allows to disable the internal PC-Speaker
1376 support, saving some memory.
1380 bool "Enable full-sized data structures for core" if EXPERT
1382 Disabling this option reduces the size of miscellaneous core
1383 kernel data structures. This saves memory on small machines,
1384 but may reduce performance.
1387 bool "Enable futex support" if EXPERT
1391 Disabling this option will cause the kernel to be built without
1392 support for "fast userspace mutexes". The resulting kernel may not
1393 run glibc-based applications correctly.
1396 bool "Enable eventpoll support" if EXPERT
1400 Disabling this option will cause the kernel to be built without
1401 support for epoll family of system calls.
1404 bool "Enable signalfd() system call" if EXPERT
1408 Enable the signalfd() system call that allows to receive signals
1409 on a file descriptor.
1414 bool "Enable timerfd() system call" if EXPERT
1418 Enable the timerfd() system call that allows to receive timer
1419 events on a file descriptor.
1424 bool "Enable eventfd() system call" if EXPERT
1428 Enable the eventfd() system call that allows to receive both
1429 kernel notification (ie. KAIO) or userspace notifications.
1434 bool "Use full shmem filesystem" if EXPERT
1438 The shmem is an internal filesystem used to manage shared memory.
1439 It is backed by swap and manages resource limits. It is also exported
1440 to userspace as tmpfs if TMPFS is enabled. Disabling this
1441 option replaces shmem and tmpfs with the much simpler ramfs code,
1442 which may be appropriate on small systems without swap.
1445 bool "Enable AIO support" if EXPERT
1448 This option enables POSIX asynchronous I/O which may by used
1449 by some high performance threaded applications. Disabling
1450 this option saves about 7k.
1454 bool "Enable PCI quirk workarounds" if EXPERT
1457 This enables workarounds for various PCI chipset
1458 bugs/quirks. Disable this only if your target machine is
1459 unaffected by PCI quirks.
1462 bool "Embedded system"
1465 This option should be enabled if compiling the kernel for
1466 an embedded system so certain expert options are available
1469 config HAVE_PERF_EVENTS
1472 See tools/perf/design.txt for details.
1474 config PERF_USE_VMALLOC
1477 See tools/perf/design.txt for details
1479 menu "Kernel Performance Events And Counters"
1482 bool "Kernel performance events and counters"
1483 default y if PROFILING
1484 depends on HAVE_PERF_EVENTS
1488 Enable kernel support for various performance events provided
1489 by software and hardware.
1491 Software events are supported either built-in or via the
1492 use of generic tracepoints.
1494 Most modern CPUs support performance events via performance
1495 counter registers. These registers count the number of certain
1496 types of hw events: such as instructions executed, cachemisses
1497 suffered, or branches mis-predicted - without slowing down the
1498 kernel or applications. These registers can also trigger interrupts
1499 when a threshold number of events have passed - and can thus be
1500 used to profile the code that runs on that CPU.
1502 The Linux Performance Event subsystem provides an abstraction of
1503 these software and hardware event capabilities, available via a
1504 system call and used by the "perf" utility in tools/perf/. It
1505 provides per task and per CPU counters, and it provides event
1506 capabilities on top of those.
1510 config DEBUG_PERF_USE_VMALLOC
1512 bool "Debug: use vmalloc to back perf mmap() buffers"
1513 depends on PERF_EVENTS && DEBUG_KERNEL
1514 select PERF_USE_VMALLOC
1516 Use vmalloc memory to back perf mmap() buffers.
1518 Mostly useful for debugging the vmalloc code on platforms
1519 that don't require it.
1525 config VM_EVENT_COUNTERS
1527 bool "Enable VM event counters for /proc/vmstat" if EXPERT
1529 VM event counters are needed for event counts to be shown.
1530 This option allows the disabling of the VM event counters
1531 on EXPERT systems. /proc/vmstat will only show page counts
1532 if VM event counters are disabled.
1536 bool "Enable SLUB debugging support" if EXPERT
1537 depends on SLUB && SYSFS
1539 SLUB has extensive debug support features. Disabling these can
1540 result in significant savings in code size. This also disables
1541 SLUB sysfs support. /sys/slab will not exist and there will be
1542 no support for cache validation etc.
1545 bool "Disable heap randomization"
1548 Randomizing heap placement makes heap exploits harder, but it
1549 also breaks ancient binaries (including anything libc5 based).
1550 This option changes the bootup default to heap randomization
1551 disabled, and can be overridden at runtime by setting
1552 /proc/sys/kernel/randomize_va_space to 2.
1554 On non-ancient distros (post-2000 ones) N is usually a safe choice.
1557 prompt "Choose SLAB allocator"
1560 This option allows to select a slab allocator.
1565 The regular slab allocator that is established and known to work
1566 well in all environments. It organizes cache hot objects in
1567 per cpu and per node queues.
1570 bool "SLUB (Unqueued Allocator)"
1572 SLUB is a slab allocator that minimizes cache line usage
1573 instead of managing queues of cached objects (SLAB approach).
1574 Per cpu caching is realized using slabs of objects instead
1575 of queues of objects. SLUB can use memory efficiently
1576 and has enhanced diagnostics. SLUB is the default choice for
1581 bool "SLOB (Simple Allocator)"
1583 SLOB replaces the stock allocator with a drastically simpler
1584 allocator. SLOB is generally more space efficient but
1585 does not perform as well on large systems.
1589 config SLUB_CPU_PARTIAL
1591 depends on SLUB && SMP
1592 bool "SLUB per cpu partial cache"
1594 Per cpu partial caches accellerate objects allocation and freeing
1595 that is local to a processor at the price of more indeterminism
1596 in the latency of the free. On overflow these caches will be cleared
1597 which requires the taking of locks that may cause latency spikes.
1598 Typically one would choose no for a realtime system.
1600 config MMAP_ALLOW_UNINITIALIZED
1601 bool "Allow mmapped anonymous memory to be uninitialized"
1602 depends on EXPERT && !MMU
1605 Normally, and according to the Linux spec, anonymous memory obtained
1606 from mmap() has it's contents cleared before it is passed to
1607 userspace. Enabling this config option allows you to request that
1608 mmap() skip that if it is given an MAP_UNINITIALIZED flag, thus
1609 providing a huge performance boost. If this option is not enabled,
1610 then the flag will be ignored.
1612 This is taken advantage of by uClibc's malloc(), and also by
1613 ELF-FDPIC binfmt's brk and stack allocator.
1615 Because of the obvious security issues, this option should only be
1616 enabled on embedded devices where you control what is run in
1617 userspace. Since that isn't generally a problem on no-MMU systems,
1618 it is normally safe to say Y here.
1620 See Documentation/nommu-mmap.txt for more information.
1623 bool "Profiling support"
1625 Say Y here to enable the extended profiling support mechanisms used
1626 by profilers such as OProfile.
1629 # Place an empty function call at each tracepoint site. Can be
1630 # dynamically changed for a probe function.
1635 source "arch/Kconfig"
1637 endmenu # General setup
1639 config HAVE_GENERIC_DMA_COHERENT
1646 depends on SLAB || SLUB_DEBUG
1654 default 0 if BASE_FULL
1655 default 1 if !BASE_FULL
1658 bool "Enable loadable module support"
1661 Kernel modules are small pieces of compiled code which can
1662 be inserted in the running kernel, rather than being
1663 permanently built into the kernel. You use the "modprobe"
1664 tool to add (and sometimes remove) them. If you say Y here,
1665 many parts of the kernel can be built as modules (by
1666 answering M instead of Y where indicated): this is most
1667 useful for infrequently used options which are not required
1668 for booting. For more information, see the man pages for
1669 modprobe, lsmod, modinfo, insmod and rmmod.
1671 If you say Y here, you will need to run "make
1672 modules_install" to put the modules under /lib/modules/
1673 where modprobe can find them (you may need to be root to do
1680 config MODULE_FORCE_LOAD
1681 bool "Forced module loading"
1684 Allow loading of modules without version information (ie. modprobe
1685 --force). Forced module loading sets the 'F' (forced) taint flag and
1686 is usually a really bad idea.
1688 config MODULE_UNLOAD
1689 bool "Module unloading"
1691 Without this option you will not be able to unload any
1692 modules (note that some modules may not be unloadable
1693 anyway), which makes your kernel smaller, faster
1694 and simpler. If unsure, say Y.
1696 config MODULE_FORCE_UNLOAD
1697 bool "Forced module unloading"
1698 depends on MODULE_UNLOAD
1700 This option allows you to force a module to unload, even if the
1701 kernel believes it is unsafe: the kernel will remove the module
1702 without waiting for anyone to stop using it (using the -f option to
1703 rmmod). This is mainly for kernel developers and desperate users.
1707 bool "Module versioning support"
1709 Usually, you have to use modules compiled with your kernel.
1710 Saying Y here makes it sometimes possible to use modules
1711 compiled for different kernels, by adding enough information
1712 to the modules to (hopefully) spot any changes which would
1713 make them incompatible with the kernel you are running. If
1716 config MODULE_SRCVERSION_ALL
1717 bool "Source checksum for all modules"
1719 Modules which contain a MODULE_VERSION get an extra "srcversion"
1720 field inserted into their modinfo section, which contains a
1721 sum of the source files which made it. This helps maintainers
1722 see exactly which source was used to build a module (since
1723 others sometimes change the module source without updating
1724 the version). With this option, such a "srcversion" field
1725 will be created for all modules. If unsure, say N.
1728 bool "Module signature verification"
1732 select ASYMMETRIC_KEY_TYPE
1733 select ASYMMETRIC_PUBLIC_KEY_SUBTYPE
1734 select PUBLIC_KEY_ALGO_RSA
1737 select X509_CERTIFICATE_PARSER
1739 Check modules for valid signatures upon load: the signature
1740 is simply appended to the module. For more information see
1741 Documentation/module-signing.txt.
1743 !!!WARNING!!! If you enable this option, you MUST make sure that the
1744 module DOES NOT get stripped after being signed. This includes the
1745 debuginfo strip done by some packagers (such as rpmbuild) and
1746 inclusion into an initramfs that wants the module size reduced.
1748 config MODULE_SIG_FORCE
1749 bool "Require modules to be validly signed"
1750 depends on MODULE_SIG
1752 Reject unsigned modules or signed modules for which we don't have a
1753 key. Without this, such modules will simply taint the kernel.
1755 config MODULE_SIG_ALL
1756 bool "Automatically sign all modules"
1758 depends on MODULE_SIG
1760 Sign all modules during make modules_install. Without this option,
1761 modules must be signed manually, using the scripts/sign-file tool.
1763 comment "Do not forget to sign required modules with scripts/sign-file"
1764 depends on MODULE_SIG_FORCE && !MODULE_SIG_ALL
1767 prompt "Which hash algorithm should modules be signed with?"
1768 depends on MODULE_SIG
1770 This determines which sort of hashing algorithm will be used during
1771 signature generation. This algorithm _must_ be built into the kernel
1772 directly so that signature verification can take place. It is not
1773 possible to load a signed module containing the algorithm to check
1774 the signature on that module.
1776 config MODULE_SIG_SHA1
1777 bool "Sign modules with SHA-1"
1780 config MODULE_SIG_SHA224
1781 bool "Sign modules with SHA-224"
1782 select CRYPTO_SHA256
1784 config MODULE_SIG_SHA256
1785 bool "Sign modules with SHA-256"
1786 select CRYPTO_SHA256
1788 config MODULE_SIG_SHA384
1789 bool "Sign modules with SHA-384"
1790 select CRYPTO_SHA512
1792 config MODULE_SIG_SHA512
1793 bool "Sign modules with SHA-512"
1794 select CRYPTO_SHA512
1798 config MODULE_SIG_HASH
1800 depends on MODULE_SIG
1801 default "sha1" if MODULE_SIG_SHA1
1802 default "sha224" if MODULE_SIG_SHA224
1803 default "sha256" if MODULE_SIG_SHA256
1804 default "sha384" if MODULE_SIG_SHA384
1805 default "sha512" if MODULE_SIG_SHA512
1809 config INIT_ALL_POSSIBLE
1812 Back when each arch used to define their own cpu_online_mask and
1813 cpu_possible_mask, some of them chose to initialize cpu_possible_mask
1814 with all 1s, and others with all 0s. When they were centralised,
1815 it was better to provide this option than to break all the archs
1816 and have several arch maintainers pursuing me down dark alleys.
1821 depends on (SMP && MODULE_UNLOAD) || HOTPLUG_CPU
1823 Need stop_machine() primitive.
1825 source "block/Kconfig"
1827 config PREEMPT_NOTIFIERS
1834 # Can be selected by architectures with broken toolchains
1835 # that get confused by correct const<->read_only section
1837 config BROKEN_RODATA
1843 Build a simple ASN.1 grammar compiler that produces a bytecode output
1844 that can be interpreted by the ASN.1 stream decoder and used to
1845 inform it as to what tags are to be expected in a stream and what
1846 functions to call on what tags.
1848 source "kernel/Kconfig.locks"