blob: 2888024e0b0abea6b6f37f4531ad0f8dfec5c7b3 [file] [log] [blame]
Dave Hansene1785e82005-06-23 00:07:49 -07001config SELECT_MEMORY_MODEL
2 def_bool y
Kees Cooka8826ee2013-01-16 18:54:17 -08003 depends on ARCH_SELECT_MEMORY_MODEL
Dave Hansene1785e82005-06-23 00:07:49 -07004
Dave Hansen3a9da762005-06-23 00:07:42 -07005choice
6 prompt "Memory model"
Dave Hansene1785e82005-06-23 00:07:49 -07007 depends on SELECT_MEMORY_MODEL
8 default DISCONTIGMEM_MANUAL if ARCH_DISCONTIGMEM_DEFAULT
Andy Whitcroftd41dee32005-06-23 00:07:54 -07009 default SPARSEMEM_MANUAL if ARCH_SPARSEMEM_DEFAULT
Dave Hansene1785e82005-06-23 00:07:49 -070010 default FLATMEM_MANUAL
Dave Hansen3a9da762005-06-23 00:07:42 -070011
Dave Hansene1785e82005-06-23 00:07:49 -070012config FLATMEM_MANUAL
Dave Hansen3a9da762005-06-23 00:07:42 -070013 bool "Flat Memory"
Anton Blanchardc898ec12006-01-06 00:12:07 -080014 depends on !(ARCH_DISCONTIGMEM_ENABLE || ARCH_SPARSEMEM_ENABLE) || ARCH_FLATMEM_ENABLE
Dave Hansen3a9da762005-06-23 00:07:42 -070015 help
16 This option allows you to change some of the ways that
17 Linux manages its memory internally. Most users will
18 only have one option here: FLATMEM. This is normal
19 and a correct option.
20
Andy Whitcroftd41dee32005-06-23 00:07:54 -070021 Some users of more advanced features like NUMA and
22 memory hotplug may have different options here.
Geert Uytterhoeven18f65332013-09-15 12:01:33 +020023 DISCONTIGMEM is a more mature, better tested system,
Andy Whitcroftd41dee32005-06-23 00:07:54 -070024 but is incompatible with memory hotplug and may suffer
25 decreased performance over SPARSEMEM. If unsure between
26 "Sparse Memory" and "Discontiguous Memory", choose
27 "Discontiguous Memory".
28
29 If unsure, choose this option (Flat Memory) over any other.
Dave Hansen3a9da762005-06-23 00:07:42 -070030
Dave Hansene1785e82005-06-23 00:07:49 -070031config DISCONTIGMEM_MANUAL
Dave Hansenf3519f92005-09-16 19:27:54 -070032 bool "Discontiguous Memory"
Dave Hansen3a9da762005-06-23 00:07:42 -070033 depends on ARCH_DISCONTIGMEM_ENABLE
34 help
Dave Hansen785dcd42005-06-23 00:07:50 -070035 This option provides enhanced support for discontiguous
36 memory systems, over FLATMEM. These systems have holes
37 in their physical address spaces, and this option provides
38 more efficient handling of these holes. However, the vast
39 majority of hardware has quite flat address spaces, and
Philipp Marekad3d0a32007-10-20 02:46:58 +020040 can have degraded performance from the extra overhead that
Dave Hansen785dcd42005-06-23 00:07:50 -070041 this option imposes.
42
43 Many NUMA configurations will have this as the only option.
44
Dave Hansen3a9da762005-06-23 00:07:42 -070045 If unsure, choose "Flat Memory" over this option.
46
Andy Whitcroftd41dee32005-06-23 00:07:54 -070047config SPARSEMEM_MANUAL
48 bool "Sparse Memory"
49 depends on ARCH_SPARSEMEM_ENABLE
50 help
51 This will be the only option for some systems, including
52 memory hotplug systems. This is normal.
53
54 For many other systems, this will be an alternative to
Dave Hansenf3519f92005-09-16 19:27:54 -070055 "Discontiguous Memory". This option provides some potential
Andy Whitcroftd41dee32005-06-23 00:07:54 -070056 performance benefits, along with decreased code complexity,
57 but it is newer, and more experimental.
58
59 If unsure, choose "Discontiguous Memory" or "Flat Memory"
60 over this option.
61
Dave Hansen3a9da762005-06-23 00:07:42 -070062endchoice
63
Dave Hansene1785e82005-06-23 00:07:49 -070064config DISCONTIGMEM
65 def_bool y
66 depends on (!SELECT_MEMORY_MODEL && ARCH_DISCONTIGMEM_ENABLE) || DISCONTIGMEM_MANUAL
67
Andy Whitcroftd41dee32005-06-23 00:07:54 -070068config SPARSEMEM
69 def_bool y
Russell King1a83e172009-10-26 16:50:12 -070070 depends on (!SELECT_MEMORY_MODEL && ARCH_SPARSEMEM_ENABLE) || SPARSEMEM_MANUAL
Andy Whitcroftd41dee32005-06-23 00:07:54 -070071
Dave Hansene1785e82005-06-23 00:07:49 -070072config FLATMEM
73 def_bool y
Andy Whitcroftd41dee32005-06-23 00:07:54 -070074 depends on (!DISCONTIGMEM && !SPARSEMEM) || FLATMEM_MANUAL
75
76config FLAT_NODE_MEM_MAP
77 def_bool y
78 depends on !SPARSEMEM
Dave Hansene1785e82005-06-23 00:07:49 -070079
Dave Hansen93b75042005-06-23 00:07:47 -070080#
81# Both the NUMA code and DISCONTIGMEM use arrays of pg_data_t's
82# to represent different areas of memory. This variable allows
83# those dependencies to exist individually.
84#
85config NEED_MULTIPLE_NODES
86 def_bool y
87 depends on DISCONTIGMEM || NUMA
Andy Whitcroftaf705362005-06-23 00:07:53 -070088
89config HAVE_MEMORY_PRESENT
90 def_bool y
Andy Whitcroftd41dee32005-06-23 00:07:54 -070091 depends on ARCH_HAVE_MEMORY_PRESENT || SPARSEMEM
Bob Picco802f1922005-09-03 15:54:26 -070092
93#
Bob Picco3e347262005-09-03 15:54:28 -070094# SPARSEMEM_EXTREME (which is the default) does some bootmem
Matt LaPlante84eb8d02006-10-03 22:53:09 +020095# allocations when memory_present() is called. If this cannot
Bob Picco3e347262005-09-03 15:54:28 -070096# be done on your architecture, select this option. However,
97# statically allocating the mem_section[] array can potentially
98# consume vast quantities of .bss, so be careful.
99#
100# This option will also potentially produce smaller runtime code
101# with gcc 3.4 and later.
102#
103config SPARSEMEM_STATIC
Jan Beulich9ba16082008-10-15 22:01:38 -0700104 bool
Bob Picco3e347262005-09-03 15:54:28 -0700105
106#
Matt LaPlante44c09202006-10-03 22:34:14 +0200107# Architecture platforms which require a two level mem_section in SPARSEMEM
Bob Picco802f1922005-09-03 15:54:26 -0700108# must select this option. This is usually for architecture platforms with
109# an extremely sparse physical address space.
110#
Bob Picco3e347262005-09-03 15:54:28 -0700111config SPARSEMEM_EXTREME
112 def_bool y
113 depends on SPARSEMEM && !SPARSEMEM_STATIC
Hugh Dickins4c21e2f2005-10-29 18:16:40 -0700114
Andy Whitcroft29c71112007-10-16 01:24:14 -0700115config SPARSEMEM_VMEMMAP_ENABLE
Jan Beulich9ba16082008-10-15 22:01:38 -0700116 bool
Andy Whitcroft29c71112007-10-16 01:24:14 -0700117
Yinghai Lu9bdac912010-02-10 01:20:22 -0800118config SPARSEMEM_ALLOC_MEM_MAP_TOGETHER
119 def_bool y
120 depends on SPARSEMEM && X86_64
121
Andy Whitcroft29c71112007-10-16 01:24:14 -0700122config SPARSEMEM_VMEMMAP
Geoff Levanda5ee6da2007-12-17 16:19:53 -0800123 bool "Sparse Memory virtual memmap"
124 depends on SPARSEMEM && SPARSEMEM_VMEMMAP_ENABLE
125 default y
126 help
127 SPARSEMEM_VMEMMAP uses a virtually mapped memmap to optimise
128 pfn_to_page and page_to_pfn operations. This is the most
129 efficient option when sufficient kernel resources are available.
Andy Whitcroft29c71112007-10-16 01:24:14 -0700130
Yinghai Lu95f72d12010-07-12 14:36:09 +1000131config HAVE_MEMBLOCK
132 boolean
133
Tejun Heo7c0caeb2011-07-14 11:43:42 +0200134config HAVE_MEMBLOCK_NODE_MAP
135 boolean
136
Tejun Heoc378ddd2011-07-14 11:46:03 +0200137config ARCH_DISCARD_MEMBLOCK
138 boolean
139
Sam Ravnborg66616722011-10-31 17:08:20 -0700140config NO_BOOTMEM
141 boolean
142
Minchan Kimee6f5092012-07-31 16:43:50 -0700143config MEMORY_ISOLATION
144 boolean
145
Lai Jiangshan20b2f522012-12-12 13:52:00 -0800146config MOVABLE_NODE
147 boolean "Enable to assign a node which has only movable memory"
148 depends on HAVE_MEMBLOCK
149 depends on NO_BOOTMEM
150 depends on X86_64
151 depends on NUMA
Tang Chenc2974052012-12-18 14:21:33 -0800152 default n
153 help
154 Allow a node to have only movable memory. Pages used by the kernel,
155 such as direct mapping pages cannot be migrated. So the corresponding
Tang Chenc5320922013-11-12 15:08:10 -0800156 memory device cannot be hotplugged. This option allows the following
157 two things:
158 - When the system is booting, node full of hotpluggable memory can
159 be arranged to have only movable memory so that the whole node can
160 be hot-removed. (need movable_node boot option specified).
161 - After the system is up, the option allows users to online all the
162 memory of a node as movable memory so that the whole node can be
163 hot-removed.
164
165 Users who don't use the memory hotplug feature are fine with this
166 option on since they don't specify movable_node boot option or they
167 don't online memory as movable.
Tang Chenc2974052012-12-18 14:21:33 -0800168
169 Say Y here if you want to hotplug a whole node.
170 Say N here if you want kernel to use memory on all nodes evenly.
Lai Jiangshan20b2f522012-12-12 13:52:00 -0800171
Yasuaki Ishimatsu46723bf2013-02-22 16:33:00 -0800172#
173# Only be set on architectures that have completely implemented memory hotplug
174# feature. If you are not sure, don't touch it.
175#
176config HAVE_BOOTMEM_INFO_NODE
177 def_bool n
178
Dave Hansen3947be12005-10-29 18:16:54 -0700179# eventually, we can have this option just 'select SPARSEMEM'
180config MEMORY_HOTPLUG
181 bool "Allow for memory hot-add"
Keith Manntheyec69acb2006-09-30 23:27:05 -0700182 depends on SPARSEMEM || X86_64_ACPI_NUMA
Stephen Rothwell40b31362013-05-21 13:49:35 +1000183 depends on ARCH_ENABLE_MEMORY_HOTPLUG
Kumar Galaed84a072009-10-16 07:21:36 +0000184 depends on (IA64 || X86 || PPC_BOOK3S_64 || SUPERH || S390)
Dave Hansen3947be12005-10-29 18:16:54 -0700185
Keith Manntheyec69acb2006-09-30 23:27:05 -0700186config MEMORY_HOTPLUG_SPARSE
187 def_bool y
188 depends on SPARSEMEM && MEMORY_HOTPLUG
189
KAMEZAWA Hiroyuki0c0e6192007-10-16 01:26:12 -0700190config MEMORY_HOTREMOVE
191 bool "Allow for memory hot remove"
Yasuaki Ishimatsu46723bf2013-02-22 16:33:00 -0800192 select MEMORY_ISOLATION
Nathan Fontenotf7e33342013-09-27 10:18:09 -0500193 select HAVE_BOOTMEM_INFO_NODE if (X86_64 || PPC64)
KAMEZAWA Hiroyuki0c0e6192007-10-16 01:26:12 -0700194 depends on MEMORY_HOTPLUG && ARCH_ENABLE_MEMORY_HOTREMOVE
195 depends on MIGRATION
196
Christoph Lametere20b8cc2008-04-28 02:12:55 -0700197#
198# If we have space for more page flags then we can enable additional
199# optimizations and functionality.
200#
201# Regular Sparsemem takes page flag bits for the sectionid if it does not
202# use a virtual memmap. Disable extended page flags for 32 bit platforms
203# that require the use of a sectionid in the page flags.
204#
205config PAGEFLAGS_EXTENDED
206 def_bool y
H. Peter Anvina269cca2009-08-31 11:17:44 -0700207 depends on 64BIT || SPARSEMEM_VMEMMAP || !SPARSEMEM
Christoph Lametere20b8cc2008-04-28 02:12:55 -0700208
Hugh Dickins4c21e2f2005-10-29 18:16:40 -0700209# Heavily threaded applications may benefit from splitting the mm-wide
210# page_table_lock, so that faults on different parts of the user address
211# space can be handled with less contention: split it at this NR_CPUS.
212# Default to 4 for wider testing, though 8 might be more appropriate.
213# ARM's adjust_pte (unused if VIPT) depends on mm-wide page_table_lock.
Hugh Dickins7b6ac9d2005-11-23 13:37:37 -0800214# PA-RISC 7xxx's spinlock_t would enlarge struct page from 32 to 44 bytes.
Hugh Dickinsa70caa82009-12-14 17:59:02 -0800215# DEBUG_SPINLOCK and DEBUG_LOCK_ALLOC spinlock_t also enlarge struct page.
Hugh Dickins4c21e2f2005-10-29 18:16:40 -0700216#
217config SPLIT_PTLOCK_CPUS
218 int
Hugh Dickinsa70caa82009-12-14 17:59:02 -0800219 default "999999" if ARM && !CPU_CACHE_VIPT
220 default "999999" if PARISC && !PA20
Hugh Dickins4c21e2f2005-10-29 18:16:40 -0700221 default "4"
Christoph Lameter7cbe34c2006-01-08 01:00:49 -0800222
Kirill A. Shutemove009bb32013-11-14 14:31:07 -0800223config ARCH_ENABLE_SPLIT_PMD_PTLOCK
224 boolean
225
Christoph Lameter7cbe34c2006-01-08 01:00:49 -0800226#
Rafael Aquini18468d92012-12-11 16:02:38 -0800227# support for memory balloon compaction
228config BALLOON_COMPACTION
229 bool "Allow for balloon memory compaction/migration"
230 def_bool y
231 depends on COMPACTION && VIRTIO_BALLOON
232 help
233 Memory fragmentation introduced by ballooning might reduce
234 significantly the number of 2MB contiguous memory blocks that can be
235 used within a guest, thus imposing performance penalties associated
236 with the reduced number of transparent huge pages that could be used
237 by the guest workload. Allowing the compaction & migration for memory
238 pages enlisted as being part of memory balloon devices avoids the
239 scenario aforementioned and helps improving memory defragmentation.
240
241#
Mel Gormane9e96b32010-05-24 14:32:21 -0700242# support for memory compaction
243config COMPACTION
244 bool "Allow for memory compaction"
Rik van Riel05106e62012-10-08 16:33:03 -0700245 def_bool y
Mel Gormane9e96b32010-05-24 14:32:21 -0700246 select MIGRATION
Andrea Arcangeli33a93872011-01-25 15:07:25 -0800247 depends on MMU
Mel Gormane9e96b32010-05-24 14:32:21 -0700248 help
249 Allows the compaction of memory for the allocation of huge pages.
250
251#
Christoph Lameter7cbe34c2006-01-08 01:00:49 -0800252# support for page migration
253#
254config MIGRATION
Christoph Lameterb20a3502006-03-22 00:09:12 -0800255 bool "Page migration"
Christoph Lameter6c5240a2006-06-23 02:03:37 -0700256 def_bool y
Chen Gangde32a812013-09-12 15:14:08 -0700257 depends on (NUMA || ARCH_ENABLE_MEMORY_HOTREMOVE || COMPACTION || CMA) && MMU
Christoph Lameterb20a3502006-03-22 00:09:12 -0800258 help
259 Allows the migration of the physical location of pages of processes
Mel Gormane9e96b32010-05-24 14:32:21 -0700260 while the virtual addresses are not changed. This is useful in
261 two situations. The first is on NUMA systems to put pages nearer
262 to the processors accessing. The second is when allocating huge
263 pages as migration can relocate pages to satisfy a huge page
264 allocation instead of reclaiming.
Greg Kroah-Hartman6550e072006-06-12 17:11:31 -0700265
Jeremy Fitzhardinge600715d2008-09-11 01:31:45 -0700266config PHYS_ADDR_T_64BIT
267 def_bool 64BIT || ARCH_PHYS_ADDR_T_64BIT
268
Christoph Lameter4b51d662007-02-10 01:43:10 -0800269config ZONE_DMA_FLAG
270 int
271 default "0" if !ZONE_DMA
272 default "1"
273
Christoph Lameter2a7326b2007-07-17 04:03:37 -0700274config BOUNCE
Vinayak Menon9ca24e22013-04-29 15:08:55 -0700275 bool "Enable bounce buffers"
276 default y
Christoph Lameter2a7326b2007-07-17 04:03:37 -0700277 depends on BLOCK && MMU && (ZONE_DMA || HIGHMEM)
Vinayak Menon9ca24e22013-04-29 15:08:55 -0700278 help
279 Enable bounce buffers for devices that cannot access
280 the full range of memory available to the CPU. Enabled
281 by default when ZONE_DMA or HIGHMEM is selected, but you
282 may say n to override this.
Christoph Lameter2a7326b2007-07-17 04:03:37 -0700283
Darrick J. Wongffecfd12013-02-21 16:42:55 -0800284# On the 'tile' arch, USB OHCI needs the bounce pool since tilegx will often
285# have more than 4GB of memory, but we don't currently use the IOTLB to present
286# a 32-bit address to OHCI. So we need to use a bounce pool instead.
287#
288# We also use the bounce pool to provide stable page writes for jbd. jbd
289# initiates buffer writeback without locking the page or setting PG_writeback,
290# and fixing that behavior (a second time; jbd2 doesn't have this problem) is
291# a major rework effort. Instead, use the bounce buffer to snapshot pages
292# (until jbd goes away). The only jbd user is ext3.
293config NEED_BOUNCE_POOL
294 bool
295 default y if (TILE && USB_OHCI_HCD) || (BLK_DEV_INTEGRITY && JBD)
296
Christoph Lameter6225e932007-05-06 14:49:50 -0700297config NR_QUICK
298 int
299 depends on QUICKLIST
Paul Mundt0176bd32010-01-05 12:35:00 +0900300 default "2" if AVR32
Christoph Lameter6225e932007-05-06 14:49:50 -0700301 default "1"
Stephen Rothwellf057eac2007-07-15 23:40:05 -0700302
303config VIRT_TO_BUS
Stephen Rothwell4febd952013-03-07 15:48:16 +1100304 bool
305 help
306 An architecture should select this if it implements the
307 deprecated interface virt_to_bus(). All new architectures
308 should probably not select this.
309
Andrea Arcangelicddb8a52008-07-28 15:46:29 -0700310
311config MMU_NOTIFIER
312 bool
David Howellsfc4d5c22009-05-06 16:03:05 -0700313
Hugh Dickinsf8af4da2009-09-21 17:01:57 -0700314config KSM
315 bool "Enable KSM for page merging"
316 depends on MMU
317 help
318 Enable Kernel Samepage Merging: KSM periodically scans those areas
319 of an application's address space that an app has advised may be
320 mergeable. When it finds pages of identical content, it replaces
Hugh Dickinsd0f209f2009-12-14 17:59:34 -0800321 the many instances by a single page with that content, so
Hugh Dickinsf8af4da2009-09-21 17:01:57 -0700322 saving memory until one or another app needs to modify the content.
323 Recommended for use with KVM, or with other duplicative applications.
Hugh Dickinsc73602a2009-10-07 16:32:22 -0700324 See Documentation/vm/ksm.txt for more information: KSM is inactive
325 until a program has madvised that an area is MADV_MERGEABLE, and
326 root has set /sys/kernel/mm/ksm/run to 1 (if CONFIG_SYSFS is set).
Hugh Dickinsf8af4da2009-09-21 17:01:57 -0700327
Christoph Lametere0a94c22009-06-03 16:04:31 -0400328config DEFAULT_MMAP_MIN_ADDR
329 int "Low address space to protect from user allocation"
David Howells6e141542009-12-15 19:27:45 +0000330 depends on MMU
Christoph Lametere0a94c22009-06-03 16:04:31 -0400331 default 4096
332 help
333 This is the portion of low virtual memory which should be protected
334 from userspace allocation. Keeping a user from writing to low pages
335 can help reduce the impact of kernel NULL pointer bugs.
336
337 For most ia64, ppc64 and x86 users with lots of address space
338 a value of 65536 is reasonable and should cause no problems.
339 On arm and other archs it should not be higher than 32768.
Eric Paris788084a2009-07-31 12:54:11 -0400340 Programs which use vm86 functionality or have some need to map
341 this low address space will need CAP_SYS_RAWIO or disable this
342 protection by setting the value to 0.
Christoph Lametere0a94c22009-06-03 16:04:31 -0400343
344 This value can be changed after boot using the
345 /proc/sys/vm/mmap_min_addr tunable.
346
Linus Torvaldsd949f362009-09-26 09:35:07 -0700347config ARCH_SUPPORTS_MEMORY_FAILURE
348 bool
Christoph Lametere0a94c22009-06-03 16:04:31 -0400349
Andi Kleen6a460792009-09-16 11:50:15 +0200350config MEMORY_FAILURE
351 depends on MMU
Linus Torvaldsd949f362009-09-26 09:35:07 -0700352 depends on ARCH_SUPPORTS_MEMORY_FAILURE
Andi Kleen6a460792009-09-16 11:50:15 +0200353 bool "Enable recovery from hardware memory errors"
Minchan Kimee6f5092012-07-31 16:43:50 -0700354 select MEMORY_ISOLATION
Andi Kleen6a460792009-09-16 11:50:15 +0200355 help
356 Enables code to recover from some memory failures on systems
357 with MCA recovery. This allows a system to continue running
358 even when some of its memory has uncorrected errors. This requires
359 special hardware support and typically ECC memory.
360
Andi Kleencae681f2009-09-16 11:50:17 +0200361config HWPOISON_INJECT
Andi Kleen413f9ef2009-12-16 12:20:00 +0100362 tristate "HWPoison pages injector"
Andi Kleen27df5062009-12-21 19:56:42 +0100363 depends on MEMORY_FAILURE && DEBUG_KERNEL && PROC_FS
Wu Fengguang478c5ff2009-12-16 12:19:59 +0100364 select PROC_PAGE_MONITOR
Andi Kleencae681f2009-09-16 11:50:17 +0200365
David Howellsfc4d5c22009-05-06 16:03:05 -0700366config NOMMU_INITIAL_TRIM_EXCESS
367 int "Turn on mmap() excess space trimming before booting"
368 depends on !MMU
369 default 1
370 help
371 The NOMMU mmap() frequently needs to allocate large contiguous chunks
372 of memory on which to store mappings, but it can only ask the system
373 allocator for chunks in 2^N*PAGE_SIZE amounts - which is frequently
374 more than it requires. To deal with this, mmap() is able to trim off
375 the excess and return it to the allocator.
376
377 If trimming is enabled, the excess is trimmed off and returned to the
378 system allocator, which can cause extra fragmentation, particularly
379 if there are a lot of transient processes.
380
381 If trimming is disabled, the excess is kept, but not used, which for
382 long-term mappings means that the space is wasted.
383
384 Trimming can be dynamically controlled through a sysctl option
385 (/proc/sys/vm/nr_trim_pages) which specifies the minimum number of
386 excess pages there must be before trimming should occur, or zero if
387 no trimming is to occur.
388
389 This option specifies the initial value of this option. The default
390 of 1 says that all excess pages should be trimmed.
391
392 See Documentation/nommu-mmap.txt for more information.
Tejun Heobbddff02010-09-03 18:22:48 +0200393
Andrea Arcangeli4c76d9d2011-01-13 15:46:39 -0800394config TRANSPARENT_HUGEPAGE
Andrea Arcangeli13ece882011-01-13 15:47:07 -0800395 bool "Transparent Hugepage Support"
Gerald Schaefer15626062012-10-08 16:30:04 -0700396 depends on HAVE_ARCH_TRANSPARENT_HUGEPAGE
Andrea Arcangeli5d689242011-01-13 15:47:07 -0800397 select COMPACTION
Andrea Arcangeli4c76d9d2011-01-13 15:46:39 -0800398 help
399 Transparent Hugepages allows the kernel to use huge pages and
400 huge tlb transparently to the applications whenever possible.
401 This feature can improve computing performance to certain
402 applications by speeding up page faults during memory
403 allocation, by reducing the number of tlb misses and by speeding
404 up the pagetable walking.
405
406 If memory constrained on embedded, you may want to say N.
407
Andrea Arcangeli13ece882011-01-13 15:47:07 -0800408choice
409 prompt "Transparent Hugepage Support sysfs defaults"
410 depends on TRANSPARENT_HUGEPAGE
411 default TRANSPARENT_HUGEPAGE_ALWAYS
412 help
413 Selects the sysfs defaults for Transparent Hugepage Support.
414
415 config TRANSPARENT_HUGEPAGE_ALWAYS
416 bool "always"
417 help
418 Enabling Transparent Hugepage always, can increase the
419 memory footprint of applications without a guaranteed
420 benefit but it will work automatically for all applications.
421
422 config TRANSPARENT_HUGEPAGE_MADVISE
423 bool "madvise"
424 help
425 Enabling Transparent Hugepage madvise, will only provide a
426 performance improvement benefit to the applications using
427 madvise(MADV_HUGEPAGE) but it won't risk to increase the
428 memory footprint of applications without a guaranteed
429 benefit.
430endchoice
431
Christopher Yeoh5febcbe2012-05-29 15:06:27 -0700432config CROSS_MEMORY_ATTACH
433 bool "Cross Memory Support"
434 depends on MMU
435 default y
436 help
437 Enabling this option adds the system calls process_vm_readv and
438 process_vm_writev which allow a process with the correct privileges
439 to directly read from or write to to another process's address space.
440 See the man page for more details.
441
Tejun Heobbddff02010-09-03 18:22:48 +0200442#
443# UP and nommu archs use km based percpu allocator
444#
445config NEED_PER_CPU_KM
446 depends on !SMP
447 bool
448 default y
Dan Magenheimer077b1f82011-05-26 10:01:36 -0600449
450config CLEANCACHE
451 bool "Enable cleancache driver to cache clean pages if tmem is present"
452 default n
453 help
454 Cleancache can be thought of as a page-granularity victim cache
455 for clean pages that the kernel's pageframe replacement algorithm
456 (PFRA) would like to keep around, but can't since there isn't enough
457 memory. So when the PFRA "evicts" a page, it first attempts to use
Michael Witten140a1ef2011-06-10 03:57:26 +0000458 cleancache code to put the data contained in that page into
Dan Magenheimer077b1f82011-05-26 10:01:36 -0600459 "transcendent memory", memory that is not directly accessible or
460 addressable by the kernel and is of unknown and possibly
461 time-varying size. And when a cleancache-enabled
462 filesystem wishes to access a page in a file on disk, it first
463 checks cleancache to see if it already contains it; if it does,
464 the page is copied into the kernel and a disk access is avoided.
465 When a transcendent memory driver is available (such as zcache or
466 Xen transcendent memory), a significant I/O reduction
467 may be achieved. When none is available, all cleancache calls
468 are reduced to a single pointer-compare-against-NULL resulting
469 in a negligible performance hit.
470
471 If unsure, say Y to enable cleancache
Dan Magenheimer27c6aec2012-04-09 17:10:34 -0600472
473config FRONTSWAP
474 bool "Enable frontswap to cache swap pages if tmem is present"
475 depends on SWAP
476 default n
477 help
478 Frontswap is so named because it can be thought of as the opposite
479 of a "backing" store for a swap device. The data is stored into
480 "transcendent memory", memory that is not directly accessible or
481 addressable by the kernel and is of unknown and possibly
482 time-varying size. When space in transcendent memory is available,
483 a significant swap I/O reduction may be achieved. When none is
484 available, all frontswap calls are reduced to a single pointer-
485 compare-against-NULL resulting in a negligible performance hit
486 and swap data is stored as normal on the matching swap device.
487
488 If unsure, say Y to enable frontswap.
Aneesh Kumar K.Vf825c732013-07-02 11:15:15 +0530489
490config CMA
491 bool "Contiguous Memory Allocator"
Chen Gangde32a812013-09-12 15:14:08 -0700492 depends on HAVE_MEMBLOCK && MMU
Aneesh Kumar K.Vf825c732013-07-02 11:15:15 +0530493 select MIGRATION
494 select MEMORY_ISOLATION
495 help
496 This enables the Contiguous Memory Allocator which allows other
497 subsystems to allocate big physically-contiguous blocks of memory.
498 CMA reserves a region of memory and allows only movable pages to
499 be allocated from it. This way, the kernel can use the memory for
500 pagecache and when a subsystem requests for contiguous area, the
501 allocated pages are migrated away to serve the contiguous request.
502
503 If unsure, say "n".
504
505config CMA_DEBUG
506 bool "CMA debug messages (DEVELOPMENT)"
507 depends on DEBUG_KERNEL && CMA
508 help
509 Turns on debug messages in CMA. This produces KERN_DEBUG
510 messages for every CMA call as well as various messages while
511 processing calls such as dma_alloc_from_contiguous().
512 This option does not affect warning and error messages.
Alexander Grafbf550fc2013-08-29 00:41:59 +0200513
Seth Jennings4e2e2772013-07-10 16:04:55 -0700514config ZBUD
515 tristate
516 default n
517 help
518 A special purpose allocator for storing compressed pages.
519 It is designed to store up to two compressed pages per physical
520 page. While this design limits storage density, it has simple and
521 deterministic reclaim properties that make it preferable to a higher
522 density approach when reclaim will be used.
523
Seth Jennings2b281112013-07-10 16:05:03 -0700524config ZSWAP
525 bool "Compressed cache for swap pages (EXPERIMENTAL)"
526 depends on FRONTSWAP && CRYPTO=y
527 select CRYPTO_LZO
528 select ZBUD
529 default n
530 help
531 A lightweight compressed cache for swap pages. It takes
532 pages that are in the process of being swapped out and attempts to
533 compress them into a dynamically allocated RAM-based memory pool.
534 This can result in a significant I/O reduction on swap device and,
535 in the case where decompressing from RAM is faster that swap device
536 reads, can also improve workload performance.
537
538 This is marked experimental because it is a new feature (as of
539 v3.11) that interacts heavily with memory reclaim. While these
540 interactions don't cause any known issues on simple memory setups,
541 they have not be fully explored on the large set of potential
542 configurations and workloads that exist.
543
Pavel Emelyanov0f8975e2013-07-03 15:01:20 -0700544config MEM_SOFT_DIRTY
545 bool "Track memory changes"
Sima Baymania844f382013-12-18 17:08:49 -0800546 depends on CHECKPOINT_RESTORE && HAVE_ARCH_SOFT_DIRTY && PROC_FS
Pavel Emelyanov0f8975e2013-07-03 15:01:20 -0700547 select PROC_PAGE_MONITOR
548 help
549 This option enables memory changes tracking by introducing a
550 soft-dirty bit on pte-s. This bit it set when someone writes
551 into a page just as regular dirty bit, but unlike the latter
552 it can be cleared by hands.
553
554 See Documentation/vm/soft-dirty.txt for more details.
Minchan Kimbcf16472014-01-30 15:45:50 -0800555
556config ZSMALLOC
557 bool "Memory allocator for compressed pages"
558 depends on MMU
559 default n
560 help
561 zsmalloc is a slab-based memory allocator designed to store
562 compressed RAM pages. zsmalloc uses virtual memory mapping
563 in order to reduce fragmentation. However, this results in a
564 non-standard allocator interface where a handle, not a pointer, is
565 returned by an alloc(). This handle must be mapped in order to
566 access the allocated space.
567
568config PGTABLE_MAPPING
569 bool "Use page table mapping to access object in zsmalloc"
570 depends on ZSMALLOC
571 help
572 By default, zsmalloc uses a copy-based object mapping method to
573 access allocations that span two pages. However, if a particular
574 architecture (ex, ARM) performs VM mapping faster than copying,
575 then you should select this. This causes zsmalloc to use page table
576 mapping rather than copying for object mapping.
577
Ben Hutchings2216ee82014-03-10 15:49:46 -0700578 You can check speed with zsmalloc benchmark:
579 https://github.com/spartacus06/zsmapbench