Loading...
1config SELECT_MEMORY_MODEL
2 def_bool y
3 depends on ARCH_SELECT_MEMORY_MODEL
4
5choice
6 prompt "Memory model"
7 depends on SELECT_MEMORY_MODEL
8 default DISCONTIGMEM_MANUAL if ARCH_DISCONTIGMEM_DEFAULT
9 default SPARSEMEM_MANUAL if ARCH_SPARSEMEM_DEFAULT
10 default FLATMEM_MANUAL
11
12config FLATMEM_MANUAL
13 bool "Flat Memory"
14 depends on !(ARCH_DISCONTIGMEM_ENABLE || ARCH_SPARSEMEM_ENABLE) || ARCH_FLATMEM_ENABLE
15 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
21 Some users of more advanced features like NUMA and
22 memory hotplug may have different options here.
23 DISCONTIGMEM is a more mature, better tested system,
24 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.
30
31config DISCONTIGMEM_MANUAL
32 bool "Discontiguous Memory"
33 depends on ARCH_DISCONTIGMEM_ENABLE
34 help
35 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
40 can have degraded performance from the extra overhead that
41 this option imposes.
42
43 Many NUMA configurations will have this as the only option.
44
45 If unsure, choose "Flat Memory" over this option.
46
47config 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
55 "Discontiguous Memory". This option provides some potential
56 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
62endchoice
63
64config DISCONTIGMEM
65 def_bool y
66 depends on (!SELECT_MEMORY_MODEL && ARCH_DISCONTIGMEM_ENABLE) || DISCONTIGMEM_MANUAL
67
68config SPARSEMEM
69 def_bool y
70 depends on (!SELECT_MEMORY_MODEL && ARCH_SPARSEMEM_ENABLE) || SPARSEMEM_MANUAL
71
72config FLATMEM
73 def_bool y
74 depends on (!DISCONTIGMEM && !SPARSEMEM) || FLATMEM_MANUAL
75
76config FLAT_NODE_MEM_MAP
77 def_bool y
78 depends on !SPARSEMEM
79
80#
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
88
89config HAVE_MEMORY_PRESENT
90 def_bool y
91 depends on ARCH_HAVE_MEMORY_PRESENT || SPARSEMEM
92
93#
94# SPARSEMEM_EXTREME (which is the default) does some bootmem
95# allocations when memory_present() is called. If this cannot
96# 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
104 bool
105
106#
107# Architecture platforms which require a two level mem_section in SPARSEMEM
108# must select this option. This is usually for architecture platforms with
109# an extremely sparse physical address space.
110#
111config SPARSEMEM_EXTREME
112 def_bool y
113 depends on SPARSEMEM && !SPARSEMEM_STATIC
114
115config SPARSEMEM_VMEMMAP_ENABLE
116 bool
117
118config SPARSEMEM_ALLOC_MEM_MAP_TOGETHER
119 def_bool y
120 depends on SPARSEMEM && X86_64
121
122config SPARSEMEM_VMEMMAP
123 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.
130
131config HAVE_MEMBLOCK
132 boolean
133
134config HAVE_MEMBLOCK_NODE_MAP
135 boolean
136
137config ARCH_DISCARD_MEMBLOCK
138 boolean
139
140config NO_BOOTMEM
141 boolean
142
143config MEMORY_ISOLATION
144 boolean
145
146config 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
152 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
156 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.
168
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.
171
172#
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
179# eventually, we can have this option just 'select SPARSEMEM'
180config MEMORY_HOTPLUG
181 bool "Allow for memory hot-add"
182 depends on SPARSEMEM || X86_64_ACPI_NUMA
183 depends on ARCH_ENABLE_MEMORY_HOTPLUG
184 depends on (IA64 || X86 || PPC_BOOK3S_64 || SUPERH || S390)
185
186config MEMORY_HOTPLUG_SPARSE
187 def_bool y
188 depends on SPARSEMEM && MEMORY_HOTPLUG
189
190config MEMORY_HOTREMOVE
191 bool "Allow for memory hot remove"
192 select MEMORY_ISOLATION
193 select HAVE_BOOTMEM_INFO_NODE if (X86_64 || PPC64)
194 depends on MEMORY_HOTPLUG && ARCH_ENABLE_MEMORY_HOTREMOVE
195 depends on MIGRATION
196
197#
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
207 depends on 64BIT || SPARSEMEM_VMEMMAP || !SPARSEMEM
208
209# 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.
214# PA-RISC 7xxx's spinlock_t would enlarge struct page from 32 to 44 bytes.
215# DEBUG_SPINLOCK and DEBUG_LOCK_ALLOC spinlock_t also enlarge struct page.
216#
217config SPLIT_PTLOCK_CPUS
218 int
219 default "999999" if !MMU
220 default "999999" if ARM && !CPU_CACHE_VIPT
221 default "999999" if PARISC && !PA20
222 default "4"
223
224config ARCH_ENABLE_SPLIT_PMD_PTLOCK
225 boolean
226
227#
228# support for memory balloon compaction
229config BALLOON_COMPACTION
230 bool "Allow for balloon memory compaction/migration"
231 def_bool y
232 depends on COMPACTION && VIRTIO_BALLOON
233 help
234 Memory fragmentation introduced by ballooning might reduce
235 significantly the number of 2MB contiguous memory blocks that can be
236 used within a guest, thus imposing performance penalties associated
237 with the reduced number of transparent huge pages that could be used
238 by the guest workload. Allowing the compaction & migration for memory
239 pages enlisted as being part of memory balloon devices avoids the
240 scenario aforementioned and helps improving memory defragmentation.
241
242#
243# support for memory compaction
244config COMPACTION
245 bool "Allow for memory compaction"
246 def_bool y
247 select MIGRATION
248 depends on MMU
249 help
250 Allows the compaction of memory for the allocation of huge pages.
251
252#
253# support for page migration
254#
255config MIGRATION
256 bool "Page migration"
257 def_bool y
258 depends on (NUMA || ARCH_ENABLE_MEMORY_HOTREMOVE || COMPACTION || CMA) && MMU
259 help
260 Allows the migration of the physical location of pages of processes
261 while the virtual addresses are not changed. This is useful in
262 two situations. The first is on NUMA systems to put pages nearer
263 to the processors accessing. The second is when allocating huge
264 pages as migration can relocate pages to satisfy a huge page
265 allocation instead of reclaiming.
266
267config PHYS_ADDR_T_64BIT
268 def_bool 64BIT || ARCH_PHYS_ADDR_T_64BIT
269
270config ZONE_DMA_FLAG
271 int
272 default "0" if !ZONE_DMA
273 default "1"
274
275config BOUNCE
276 bool "Enable bounce buffers"
277 default y
278 depends on BLOCK && MMU && (ZONE_DMA || HIGHMEM)
279 help
280 Enable bounce buffers for devices that cannot access
281 the full range of memory available to the CPU. Enabled
282 by default when ZONE_DMA or HIGHMEM is selected, but you
283 may say n to override this.
284
285# On the 'tile' arch, USB OHCI needs the bounce pool since tilegx will often
286# have more than 4GB of memory, but we don't currently use the IOTLB to present
287# a 32-bit address to OHCI. So we need to use a bounce pool instead.
288#
289# We also use the bounce pool to provide stable page writes for jbd. jbd
290# initiates buffer writeback without locking the page or setting PG_writeback,
291# and fixing that behavior (a second time; jbd2 doesn't have this problem) is
292# a major rework effort. Instead, use the bounce buffer to snapshot pages
293# (until jbd goes away). The only jbd user is ext3.
294config NEED_BOUNCE_POOL
295 bool
296 default y if (TILE && USB_OHCI_HCD) || (BLK_DEV_INTEGRITY && JBD)
297
298config NR_QUICK
299 int
300 depends on QUICKLIST
301 default "2" if AVR32
302 default "1"
303
304config VIRT_TO_BUS
305 bool
306 help
307 An architecture should select this if it implements the
308 deprecated interface virt_to_bus(). All new architectures
309 should probably not select this.
310
311
312config MMU_NOTIFIER
313 bool
314
315config KSM
316 bool "Enable KSM for page merging"
317 depends on MMU
318 help
319 Enable Kernel Samepage Merging: KSM periodically scans those areas
320 of an application's address space that an app has advised may be
321 mergeable. When it finds pages of identical content, it replaces
322 the many instances by a single page with that content, so
323 saving memory until one or another app needs to modify the content.
324 Recommended for use with KVM, or with other duplicative applications.
325 See Documentation/vm/ksm.txt for more information: KSM is inactive
326 until a program has madvised that an area is MADV_MERGEABLE, and
327 root has set /sys/kernel/mm/ksm/run to 1 (if CONFIG_SYSFS is set).
328
329config DEFAULT_MMAP_MIN_ADDR
330 int "Low address space to protect from user allocation"
331 depends on MMU
332 default 4096
333 help
334 This is the portion of low virtual memory which should be protected
335 from userspace allocation. Keeping a user from writing to low pages
336 can help reduce the impact of kernel NULL pointer bugs.
337
338 For most ia64, ppc64 and x86 users with lots of address space
339 a value of 65536 is reasonable and should cause no problems.
340 On arm and other archs it should not be higher than 32768.
341 Programs which use vm86 functionality or have some need to map
342 this low address space will need CAP_SYS_RAWIO or disable this
343 protection by setting the value to 0.
344
345 This value can be changed after boot using the
346 /proc/sys/vm/mmap_min_addr tunable.
347
348config ARCH_SUPPORTS_MEMORY_FAILURE
349 bool
350
351config MEMORY_FAILURE
352 depends on MMU
353 depends on ARCH_SUPPORTS_MEMORY_FAILURE
354 bool "Enable recovery from hardware memory errors"
355 select MEMORY_ISOLATION
356 help
357 Enables code to recover from some memory failures on systems
358 with MCA recovery. This allows a system to continue running
359 even when some of its memory has uncorrected errors. This requires
360 special hardware support and typically ECC memory.
361
362config HWPOISON_INJECT
363 tristate "HWPoison pages injector"
364 depends on MEMORY_FAILURE && DEBUG_KERNEL && PROC_FS
365 select PROC_PAGE_MONITOR
366
367config NOMMU_INITIAL_TRIM_EXCESS
368 int "Turn on mmap() excess space trimming before booting"
369 depends on !MMU
370 default 1
371 help
372 The NOMMU mmap() frequently needs to allocate large contiguous chunks
373 of memory on which to store mappings, but it can only ask the system
374 allocator for chunks in 2^N*PAGE_SIZE amounts - which is frequently
375 more than it requires. To deal with this, mmap() is able to trim off
376 the excess and return it to the allocator.
377
378 If trimming is enabled, the excess is trimmed off and returned to the
379 system allocator, which can cause extra fragmentation, particularly
380 if there are a lot of transient processes.
381
382 If trimming is disabled, the excess is kept, but not used, which for
383 long-term mappings means that the space is wasted.
384
385 Trimming can be dynamically controlled through a sysctl option
386 (/proc/sys/vm/nr_trim_pages) which specifies the minimum number of
387 excess pages there must be before trimming should occur, or zero if
388 no trimming is to occur.
389
390 This option specifies the initial value of this option. The default
391 of 1 says that all excess pages should be trimmed.
392
393 See Documentation/nommu-mmap.txt for more information.
394
395config TRANSPARENT_HUGEPAGE
396 bool "Transparent Hugepage Support"
397 depends on HAVE_ARCH_TRANSPARENT_HUGEPAGE
398 select COMPACTION
399 help
400 Transparent Hugepages allows the kernel to use huge pages and
401 huge tlb transparently to the applications whenever possible.
402 This feature can improve computing performance to certain
403 applications by speeding up page faults during memory
404 allocation, by reducing the number of tlb misses and by speeding
405 up the pagetable walking.
406
407 If memory constrained on embedded, you may want to say N.
408
409choice
410 prompt "Transparent Hugepage Support sysfs defaults"
411 depends on TRANSPARENT_HUGEPAGE
412 default TRANSPARENT_HUGEPAGE_ALWAYS
413 help
414 Selects the sysfs defaults for Transparent Hugepage Support.
415
416 config TRANSPARENT_HUGEPAGE_ALWAYS
417 bool "always"
418 help
419 Enabling Transparent Hugepage always, can increase the
420 memory footprint of applications without a guaranteed
421 benefit but it will work automatically for all applications.
422
423 config TRANSPARENT_HUGEPAGE_MADVISE
424 bool "madvise"
425 help
426 Enabling Transparent Hugepage madvise, will only provide a
427 performance improvement benefit to the applications using
428 madvise(MADV_HUGEPAGE) but it won't risk to increase the
429 memory footprint of applications without a guaranteed
430 benefit.
431endchoice
432
433config CROSS_MEMORY_ATTACH
434 bool "Cross Memory Support"
435 depends on MMU
436 default y
437 help
438 Enabling this option adds the system calls process_vm_readv and
439 process_vm_writev which allow a process with the correct privileges
440 to directly read from or write to to another process's address space.
441 See the man page for more details.
442
443#
444# UP and nommu archs use km based percpu allocator
445#
446config NEED_PER_CPU_KM
447 depends on !SMP
448 bool
449 default y
450
451config CLEANCACHE
452 bool "Enable cleancache driver to cache clean pages if tmem is present"
453 default n
454 help
455 Cleancache can be thought of as a page-granularity victim cache
456 for clean pages that the kernel's pageframe replacement algorithm
457 (PFRA) would like to keep around, but can't since there isn't enough
458 memory. So when the PFRA "evicts" a page, it first attempts to use
459 cleancache code to put the data contained in that page into
460 "transcendent memory", memory that is not directly accessible or
461 addressable by the kernel and is of unknown and possibly
462 time-varying size. And when a cleancache-enabled
463 filesystem wishes to access a page in a file on disk, it first
464 checks cleancache to see if it already contains it; if it does,
465 the page is copied into the kernel and a disk access is avoided.
466 When a transcendent memory driver is available (such as zcache or
467 Xen transcendent memory), a significant I/O reduction
468 may be achieved. When none is available, all cleancache calls
469 are reduced to a single pointer-compare-against-NULL resulting
470 in a negligible performance hit.
471
472 If unsure, say Y to enable cleancache
473
474config FRONTSWAP
475 bool "Enable frontswap to cache swap pages if tmem is present"
476 depends on SWAP
477 default n
478 help
479 Frontswap is so named because it can be thought of as the opposite
480 of a "backing" store for a swap device. The data is stored into
481 "transcendent memory", memory that is not directly accessible or
482 addressable by the kernel and is of unknown and possibly
483 time-varying size. When space in transcendent memory is available,
484 a significant swap I/O reduction may be achieved. When none is
485 available, all frontswap calls are reduced to a single pointer-
486 compare-against-NULL resulting in a negligible performance hit
487 and swap data is stored as normal on the matching swap device.
488
489 If unsure, say Y to enable frontswap.
490
491config CMA
492 bool "Contiguous Memory Allocator"
493 depends on HAVE_MEMBLOCK && MMU
494 select MIGRATION
495 select MEMORY_ISOLATION
496 help
497 This enables the Contiguous Memory Allocator which allows other
498 subsystems to allocate big physically-contiguous blocks of memory.
499 CMA reserves a region of memory and allows only movable pages to
500 be allocated from it. This way, the kernel can use the memory for
501 pagecache and when a subsystem requests for contiguous area, the
502 allocated pages are migrated away to serve the contiguous request.
503
504 If unsure, say "n".
505
506config CMA_DEBUG
507 bool "CMA debug messages (DEVELOPMENT)"
508 depends on DEBUG_KERNEL && CMA
509 help
510 Turns on debug messages in CMA. This produces KERN_DEBUG
511 messages for every CMA call as well as various messages while
512 processing calls such as dma_alloc_from_contiguous().
513 This option does not affect warning and error messages.
514
515config ZBUD
516 tristate
517 default n
518 help
519 A special purpose allocator for storing compressed pages.
520 It is designed to store up to two compressed pages per physical
521 page. While this design limits storage density, it has simple and
522 deterministic reclaim properties that make it preferable to a higher
523 density approach when reclaim will be used.
524
525config ZSWAP
526 bool "Compressed cache for swap pages (EXPERIMENTAL)"
527 depends on FRONTSWAP && CRYPTO=y
528 select CRYPTO_LZO
529 select ZBUD
530 default n
531 help
532 A lightweight compressed cache for swap pages. It takes
533 pages that are in the process of being swapped out and attempts to
534 compress them into a dynamically allocated RAM-based memory pool.
535 This can result in a significant I/O reduction on swap device and,
536 in the case where decompressing from RAM is faster that swap device
537 reads, can also improve workload performance.
538
539 This is marked experimental because it is a new feature (as of
540 v3.11) that interacts heavily with memory reclaim. While these
541 interactions don't cause any known issues on simple memory setups,
542 they have not be fully explored on the large set of potential
543 configurations and workloads that exist.
544
545config MEM_SOFT_DIRTY
546 bool "Track memory changes"
547 depends on CHECKPOINT_RESTORE && HAVE_ARCH_SOFT_DIRTY && PROC_FS
548 select PROC_PAGE_MONITOR
549 help
550 This option enables memory changes tracking by introducing a
551 soft-dirty bit on pte-s. This bit it set when someone writes
552 into a page just as regular dirty bit, but unlike the latter
553 it can be cleared by hands.
554
555 See Documentation/vm/soft-dirty.txt for more details.
556
557config ZSMALLOC
558 bool "Memory allocator for compressed pages"
559 depends on MMU
560 default n
561 help
562 zsmalloc is a slab-based memory allocator designed to store
563 compressed RAM pages. zsmalloc uses virtual memory mapping
564 in order to reduce fragmentation. However, this results in a
565 non-standard allocator interface where a handle, not a pointer, is
566 returned by an alloc(). This handle must be mapped in order to
567 access the allocated space.
568
569config PGTABLE_MAPPING
570 bool "Use page table mapping to access object in zsmalloc"
571 depends on ZSMALLOC
572 help
573 By default, zsmalloc uses a copy-based object mapping method to
574 access allocations that span two pages. However, if a particular
575 architecture (ex, ARM) performs VM mapping faster than copying,
576 then you should select this. This causes zsmalloc to use page table
577 mapping rather than copying for object mapping.
578
579 You can check speed with zsmalloc benchmark:
580 https://github.com/spartacus06/zsmapbench
581
582config GENERIC_EARLY_IOREMAP
583 bool
584
585config MAX_STACK_SIZE_MB
586 int "Maximum user stack size for 32-bit processes (MB)"
587 default 80
588 range 8 256 if METAG
589 range 8 2048
590 depends on STACK_GROWSUP && (!64BIT || COMPAT)
591 help
592 This is the maximum stack size in Megabytes in the VM layout of 32-bit
593 user processes when the stack grows upwards (currently only on parisc
594 and metag arch). The stack will be located at the highest memory
595 address minus the given value, unless the RLIMIT_STACK hard limit is
596 changed to a smaller value in which case that is used.
597
598 A sane initial value is 80 MB.
1# SPDX-License-Identifier: GPL-2.0-only
2
3menu "Memory Management options"
4
5config SELECT_MEMORY_MODEL
6 def_bool y
7 depends on ARCH_SELECT_MEMORY_MODEL
8
9choice
10 prompt "Memory model"
11 depends on SELECT_MEMORY_MODEL
12 default DISCONTIGMEM_MANUAL if ARCH_DISCONTIGMEM_DEFAULT
13 default SPARSEMEM_MANUAL if ARCH_SPARSEMEM_DEFAULT
14 default FLATMEM_MANUAL
15 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 selected by the architecture
19 configuration. This is normal.
20
21config FLATMEM_MANUAL
22 bool "Flat Memory"
23 depends on !(ARCH_DISCONTIGMEM_ENABLE || ARCH_SPARSEMEM_ENABLE) || ARCH_FLATMEM_ENABLE
24 help
25 This option is best suited for non-NUMA systems with
26 flat address space. The FLATMEM is the most efficient
27 system in terms of performance and resource consumption
28 and it is the best option for smaller systems.
29
30 For systems that have holes in their physical address
31 spaces and for features like NUMA and memory hotplug,
32 choose "Sparse Memory"
33
34 If unsure, choose this option (Flat Memory) over any other.
35
36config DISCONTIGMEM_MANUAL
37 bool "Discontiguous Memory"
38 depends on ARCH_DISCONTIGMEM_ENABLE
39 help
40 This option provides enhanced support for discontiguous
41 memory systems, over FLATMEM. These systems have holes
42 in their physical address spaces, and this option provides
43 more efficient handling of these holes.
44
45 Although "Discontiguous Memory" is still used by several
46 architectures, it is considered deprecated in favor of
47 "Sparse Memory".
48
49 If unsure, choose "Sparse Memory" over this option.
50
51config SPARSEMEM_MANUAL
52 bool "Sparse Memory"
53 depends on ARCH_SPARSEMEM_ENABLE
54 help
55 This will be the only option for some systems, including
56 memory hot-plug systems. This is normal.
57
58 This option provides efficient support for systems with
59 holes is their physical address space and allows memory
60 hot-plug and hot-remove.
61
62 If unsure, choose "Flat Memory" over this option.
63
64endchoice
65
66config DISCONTIGMEM
67 def_bool y
68 depends on (!SELECT_MEMORY_MODEL && ARCH_DISCONTIGMEM_ENABLE) || DISCONTIGMEM_MANUAL
69
70config SPARSEMEM
71 def_bool y
72 depends on (!SELECT_MEMORY_MODEL && ARCH_SPARSEMEM_ENABLE) || SPARSEMEM_MANUAL
73
74config FLATMEM
75 def_bool y
76 depends on (!DISCONTIGMEM && !SPARSEMEM) || FLATMEM_MANUAL
77
78config FLAT_NODE_MEM_MAP
79 def_bool y
80 depends on !SPARSEMEM
81
82#
83# Both the NUMA code and DISCONTIGMEM use arrays of pg_data_t's
84# to represent different areas of memory. This variable allows
85# those dependencies to exist individually.
86#
87config NEED_MULTIPLE_NODES
88 def_bool y
89 depends on DISCONTIGMEM || NUMA
90
91config HAVE_MEMORY_PRESENT
92 def_bool y
93 depends on ARCH_HAVE_MEMORY_PRESENT || SPARSEMEM
94
95#
96# SPARSEMEM_EXTREME (which is the default) does some bootmem
97# allocations when memory_present() is called. If this cannot
98# be done on your architecture, select this option. However,
99# statically allocating the mem_section[] array can potentially
100# consume vast quantities of .bss, so be careful.
101#
102# This option will also potentially produce smaller runtime code
103# with gcc 3.4 and later.
104#
105config SPARSEMEM_STATIC
106 bool
107
108#
109# Architecture platforms which require a two level mem_section in SPARSEMEM
110# must select this option. This is usually for architecture platforms with
111# an extremely sparse physical address space.
112#
113config SPARSEMEM_EXTREME
114 def_bool y
115 depends on SPARSEMEM && !SPARSEMEM_STATIC
116
117config SPARSEMEM_VMEMMAP_ENABLE
118 bool
119
120config SPARSEMEM_VMEMMAP
121 bool "Sparse Memory virtual memmap"
122 depends on SPARSEMEM && SPARSEMEM_VMEMMAP_ENABLE
123 default y
124 help
125 SPARSEMEM_VMEMMAP uses a virtually mapped memmap to optimise
126 pfn_to_page and page_to_pfn operations. This is the most
127 efficient option when sufficient kernel resources are available.
128
129config HAVE_MEMBLOCK_NODE_MAP
130 bool
131
132config HAVE_MEMBLOCK_PHYS_MAP
133 bool
134
135config HAVE_FAST_GUP
136 depends on MMU
137 bool
138
139config ARCH_KEEP_MEMBLOCK
140 bool
141
142config MEMORY_ISOLATION
143 bool
144
145#
146# Only be set on architectures that have completely implemented memory hotplug
147# feature. If you are not sure, don't touch it.
148#
149config HAVE_BOOTMEM_INFO_NODE
150 def_bool n
151
152# eventually, we can have this option just 'select SPARSEMEM'
153config MEMORY_HOTPLUG
154 bool "Allow for memory hot-add"
155 depends on SPARSEMEM || X86_64_ACPI_NUMA
156 depends on ARCH_ENABLE_MEMORY_HOTPLUG
157
158config MEMORY_HOTPLUG_SPARSE
159 def_bool y
160 depends on SPARSEMEM && MEMORY_HOTPLUG
161
162config MEMORY_HOTPLUG_DEFAULT_ONLINE
163 bool "Online the newly added memory blocks by default"
164 depends on MEMORY_HOTPLUG
165 help
166 This option sets the default policy setting for memory hotplug
167 onlining policy (/sys/devices/system/memory/auto_online_blocks) which
168 determines what happens to newly added memory regions. Policy setting
169 can always be changed at runtime.
170 See Documentation/admin-guide/mm/memory-hotplug.rst for more information.
171
172 Say Y here if you want all hot-plugged memory blocks to appear in
173 'online' state by default.
174 Say N here if you want the default policy to keep all hot-plugged
175 memory blocks in 'offline' state.
176
177config MEMORY_HOTREMOVE
178 bool "Allow for memory hot remove"
179 select MEMORY_ISOLATION
180 select HAVE_BOOTMEM_INFO_NODE if (X86_64 || PPC64)
181 depends on MEMORY_HOTPLUG && ARCH_ENABLE_MEMORY_HOTREMOVE
182 depends on MIGRATION
183
184# Heavily threaded applications may benefit from splitting the mm-wide
185# page_table_lock, so that faults on different parts of the user address
186# space can be handled with less contention: split it at this NR_CPUS.
187# Default to 4 for wider testing, though 8 might be more appropriate.
188# ARM's adjust_pte (unused if VIPT) depends on mm-wide page_table_lock.
189# PA-RISC 7xxx's spinlock_t would enlarge struct page from 32 to 44 bytes.
190# DEBUG_SPINLOCK and DEBUG_LOCK_ALLOC spinlock_t also enlarge struct page.
191#
192config SPLIT_PTLOCK_CPUS
193 int
194 default "999999" if !MMU
195 default "999999" if ARM && !CPU_CACHE_VIPT
196 default "999999" if PARISC && !PA20
197 default "4"
198
199config ARCH_ENABLE_SPLIT_PMD_PTLOCK
200 bool
201
202#
203# support for memory balloon
204config MEMORY_BALLOON
205 bool
206
207#
208# support for memory balloon compaction
209config BALLOON_COMPACTION
210 bool "Allow for balloon memory compaction/migration"
211 def_bool y
212 depends on COMPACTION && MEMORY_BALLOON
213 help
214 Memory fragmentation introduced by ballooning might reduce
215 significantly the number of 2MB contiguous memory blocks that can be
216 used within a guest, thus imposing performance penalties associated
217 with the reduced number of transparent huge pages that could be used
218 by the guest workload. Allowing the compaction & migration for memory
219 pages enlisted as being part of memory balloon devices avoids the
220 scenario aforementioned and helps improving memory defragmentation.
221
222#
223# support for memory compaction
224config COMPACTION
225 bool "Allow for memory compaction"
226 def_bool y
227 select MIGRATION
228 depends on MMU
229 help
230 Compaction is the only memory management component to form
231 high order (larger physically contiguous) memory blocks
232 reliably. The page allocator relies on compaction heavily and
233 the lack of the feature can lead to unexpected OOM killer
234 invocations for high order memory requests. You shouldn't
235 disable this option unless there really is a strong reason for
236 it and then we would be really interested to hear about that at
237 linux-mm@kvack.org.
238
239#
240# support for page migration
241#
242config MIGRATION
243 bool "Page migration"
244 def_bool y
245 depends on (NUMA || ARCH_ENABLE_MEMORY_HOTREMOVE || COMPACTION || CMA) && MMU
246 help
247 Allows the migration of the physical location of pages of processes
248 while the virtual addresses are not changed. This is useful in
249 two situations. The first is on NUMA systems to put pages nearer
250 to the processors accessing. The second is when allocating huge
251 pages as migration can relocate pages to satisfy a huge page
252 allocation instead of reclaiming.
253
254config ARCH_ENABLE_HUGEPAGE_MIGRATION
255 bool
256
257config ARCH_ENABLE_THP_MIGRATION
258 bool
259
260config CONTIG_ALLOC
261 def_bool (MEMORY_ISOLATION && COMPACTION) || CMA
262
263config PHYS_ADDR_T_64BIT
264 def_bool 64BIT
265
266config BOUNCE
267 bool "Enable bounce buffers"
268 default y
269 depends on BLOCK && MMU && (ZONE_DMA || HIGHMEM)
270 help
271 Enable bounce buffers for devices that cannot access
272 the full range of memory available to the CPU. Enabled
273 by default when ZONE_DMA or HIGHMEM is selected, but you
274 may say n to override this.
275
276config VIRT_TO_BUS
277 bool
278 help
279 An architecture should select this if it implements the
280 deprecated interface virt_to_bus(). All new architectures
281 should probably not select this.
282
283
284config MMU_NOTIFIER
285 bool
286 select SRCU
287
288config KSM
289 bool "Enable KSM for page merging"
290 depends on MMU
291 select XXHASH
292 help
293 Enable Kernel Samepage Merging: KSM periodically scans those areas
294 of an application's address space that an app has advised may be
295 mergeable. When it finds pages of identical content, it replaces
296 the many instances by a single page with that content, so
297 saving memory until one or another app needs to modify the content.
298 Recommended for use with KVM, or with other duplicative applications.
299 See Documentation/vm/ksm.rst for more information: KSM is inactive
300 until a program has madvised that an area is MADV_MERGEABLE, and
301 root has set /sys/kernel/mm/ksm/run to 1 (if CONFIG_SYSFS is set).
302
303config DEFAULT_MMAP_MIN_ADDR
304 int "Low address space to protect from user allocation"
305 depends on MMU
306 default 4096
307 help
308 This is the portion of low virtual memory which should be protected
309 from userspace allocation. Keeping a user from writing to low pages
310 can help reduce the impact of kernel NULL pointer bugs.
311
312 For most ia64, ppc64 and x86 users with lots of address space
313 a value of 65536 is reasonable and should cause no problems.
314 On arm and other archs it should not be higher than 32768.
315 Programs which use vm86 functionality or have some need to map
316 this low address space will need CAP_SYS_RAWIO or disable this
317 protection by setting the value to 0.
318
319 This value can be changed after boot using the
320 /proc/sys/vm/mmap_min_addr tunable.
321
322config ARCH_SUPPORTS_MEMORY_FAILURE
323 bool
324
325config MEMORY_FAILURE
326 depends on MMU
327 depends on ARCH_SUPPORTS_MEMORY_FAILURE
328 bool "Enable recovery from hardware memory errors"
329 select MEMORY_ISOLATION
330 select RAS
331 help
332 Enables code to recover from some memory failures on systems
333 with MCA recovery. This allows a system to continue running
334 even when some of its memory has uncorrected errors. This requires
335 special hardware support and typically ECC memory.
336
337config HWPOISON_INJECT
338 tristate "HWPoison pages injector"
339 depends on MEMORY_FAILURE && DEBUG_KERNEL && PROC_FS
340 select PROC_PAGE_MONITOR
341
342config NOMMU_INITIAL_TRIM_EXCESS
343 int "Turn on mmap() excess space trimming before booting"
344 depends on !MMU
345 default 1
346 help
347 The NOMMU mmap() frequently needs to allocate large contiguous chunks
348 of memory on which to store mappings, but it can only ask the system
349 allocator for chunks in 2^N*PAGE_SIZE amounts - which is frequently
350 more than it requires. To deal with this, mmap() is able to trim off
351 the excess and return it to the allocator.
352
353 If trimming is enabled, the excess is trimmed off and returned to the
354 system allocator, which can cause extra fragmentation, particularly
355 if there are a lot of transient processes.
356
357 If trimming is disabled, the excess is kept, but not used, which for
358 long-term mappings means that the space is wasted.
359
360 Trimming can be dynamically controlled through a sysctl option
361 (/proc/sys/vm/nr_trim_pages) which specifies the minimum number of
362 excess pages there must be before trimming should occur, or zero if
363 no trimming is to occur.
364
365 This option specifies the initial value of this option. The default
366 of 1 says that all excess pages should be trimmed.
367
368 See Documentation/nommu-mmap.txt for more information.
369
370config TRANSPARENT_HUGEPAGE
371 bool "Transparent Hugepage Support"
372 depends on HAVE_ARCH_TRANSPARENT_HUGEPAGE
373 select COMPACTION
374 select XARRAY_MULTI
375 help
376 Transparent Hugepages allows the kernel to use huge pages and
377 huge tlb transparently to the applications whenever possible.
378 This feature can improve computing performance to certain
379 applications by speeding up page faults during memory
380 allocation, by reducing the number of tlb misses and by speeding
381 up the pagetable walking.
382
383 If memory constrained on embedded, you may want to say N.
384
385choice
386 prompt "Transparent Hugepage Support sysfs defaults"
387 depends on TRANSPARENT_HUGEPAGE
388 default TRANSPARENT_HUGEPAGE_ALWAYS
389 help
390 Selects the sysfs defaults for Transparent Hugepage Support.
391
392 config TRANSPARENT_HUGEPAGE_ALWAYS
393 bool "always"
394 help
395 Enabling Transparent Hugepage always, can increase the
396 memory footprint of applications without a guaranteed
397 benefit but it will work automatically for all applications.
398
399 config TRANSPARENT_HUGEPAGE_MADVISE
400 bool "madvise"
401 help
402 Enabling Transparent Hugepage madvise, will only provide a
403 performance improvement benefit to the applications using
404 madvise(MADV_HUGEPAGE) but it won't risk to increase the
405 memory footprint of applications without a guaranteed
406 benefit.
407endchoice
408
409config ARCH_WANTS_THP_SWAP
410 def_bool n
411
412config THP_SWAP
413 def_bool y
414 depends on TRANSPARENT_HUGEPAGE && ARCH_WANTS_THP_SWAP && SWAP
415 help
416 Swap transparent huge pages in one piece, without splitting.
417 XXX: For now, swap cluster backing transparent huge page
418 will be split after swapout.
419
420 For selection by architectures with reasonable THP sizes.
421
422config TRANSPARENT_HUGE_PAGECACHE
423 def_bool y
424 depends on TRANSPARENT_HUGEPAGE
425
426#
427# UP and nommu archs use km based percpu allocator
428#
429config NEED_PER_CPU_KM
430 depends on !SMP
431 bool
432 default y
433
434config CLEANCACHE
435 bool "Enable cleancache driver to cache clean pages if tmem is present"
436 help
437 Cleancache can be thought of as a page-granularity victim cache
438 for clean pages that the kernel's pageframe replacement algorithm
439 (PFRA) would like to keep around, but can't since there isn't enough
440 memory. So when the PFRA "evicts" a page, it first attempts to use
441 cleancache code to put the data contained in that page into
442 "transcendent memory", memory that is not directly accessible or
443 addressable by the kernel and is of unknown and possibly
444 time-varying size. And when a cleancache-enabled
445 filesystem wishes to access a page in a file on disk, it first
446 checks cleancache to see if it already contains it; if it does,
447 the page is copied into the kernel and a disk access is avoided.
448 When a transcendent memory driver is available (such as zcache or
449 Xen transcendent memory), a significant I/O reduction
450 may be achieved. When none is available, all cleancache calls
451 are reduced to a single pointer-compare-against-NULL resulting
452 in a negligible performance hit.
453
454 If unsure, say Y to enable cleancache
455
456config FRONTSWAP
457 bool "Enable frontswap to cache swap pages if tmem is present"
458 depends on SWAP
459 help
460 Frontswap is so named because it can be thought of as the opposite
461 of a "backing" store for a swap device. The data is stored into
462 "transcendent memory", memory that is not directly accessible or
463 addressable by the kernel and is of unknown and possibly
464 time-varying size. When space in transcendent memory is available,
465 a significant swap I/O reduction may be achieved. When none is
466 available, all frontswap calls are reduced to a single pointer-
467 compare-against-NULL resulting in a negligible performance hit
468 and swap data is stored as normal on the matching swap device.
469
470 If unsure, say Y to enable frontswap.
471
472config CMA
473 bool "Contiguous Memory Allocator"
474 depends on MMU
475 select MIGRATION
476 select MEMORY_ISOLATION
477 help
478 This enables the Contiguous Memory Allocator which allows other
479 subsystems to allocate big physically-contiguous blocks of memory.
480 CMA reserves a region of memory and allows only movable pages to
481 be allocated from it. This way, the kernel can use the memory for
482 pagecache and when a subsystem requests for contiguous area, the
483 allocated pages are migrated away to serve the contiguous request.
484
485 If unsure, say "n".
486
487config CMA_DEBUG
488 bool "CMA debug messages (DEVELOPMENT)"
489 depends on DEBUG_KERNEL && CMA
490 help
491 Turns on debug messages in CMA. This produces KERN_DEBUG
492 messages for every CMA call as well as various messages while
493 processing calls such as dma_alloc_from_contiguous().
494 This option does not affect warning and error messages.
495
496config CMA_DEBUGFS
497 bool "CMA debugfs interface"
498 depends on CMA && DEBUG_FS
499 help
500 Turns on the DebugFS interface for CMA.
501
502config CMA_AREAS
503 int "Maximum count of the CMA areas"
504 depends on CMA
505 default 7
506 help
507 CMA allows to create CMA areas for particular purpose, mainly,
508 used as device private area. This parameter sets the maximum
509 number of CMA area in the system.
510
511 If unsure, leave the default value "7".
512
513config MEM_SOFT_DIRTY
514 bool "Track memory changes"
515 depends on CHECKPOINT_RESTORE && HAVE_ARCH_SOFT_DIRTY && PROC_FS
516 select PROC_PAGE_MONITOR
517 help
518 This option enables memory changes tracking by introducing a
519 soft-dirty bit on pte-s. This bit it set when someone writes
520 into a page just as regular dirty bit, but unlike the latter
521 it can be cleared by hands.
522
523 See Documentation/admin-guide/mm/soft-dirty.rst for more details.
524
525config ZSWAP
526 bool "Compressed cache for swap pages (EXPERIMENTAL)"
527 depends on FRONTSWAP && CRYPTO=y
528 select CRYPTO_LZO
529 select ZPOOL
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
544config ZPOOL
545 tristate "Common API for compressed memory storage"
546 help
547 Compressed memory storage API. This allows using either zbud or
548 zsmalloc.
549
550config ZBUD
551 tristate "Low (Up to 2x) density storage for compressed pages"
552 help
553 A special purpose allocator for storing compressed pages.
554 It is designed to store up to two compressed pages per physical
555 page. While this design limits storage density, it has simple and
556 deterministic reclaim properties that make it preferable to a higher
557 density approach when reclaim will be used.
558
559config Z3FOLD
560 tristate "Up to 3x density storage for compressed pages"
561 depends on ZPOOL
562 help
563 A special purpose allocator for storing compressed pages.
564 It is designed to store up to three compressed pages per physical
565 page. It is a ZBUD derivative so the simplicity and determinism are
566 still there.
567
568config ZSMALLOC
569 tristate "Memory allocator for compressed pages"
570 depends on MMU
571 help
572 zsmalloc is a slab-based memory allocator designed to store
573 compressed RAM pages. zsmalloc uses virtual memory mapping
574 in order to reduce fragmentation. However, this results in a
575 non-standard allocator interface where a handle, not a pointer, is
576 returned by an alloc(). This handle must be mapped in order to
577 access the allocated space.
578
579config PGTABLE_MAPPING
580 bool "Use page table mapping to access object in zsmalloc"
581 depends on ZSMALLOC
582 help
583 By default, zsmalloc uses a copy-based object mapping method to
584 access allocations that span two pages. However, if a particular
585 architecture (ex, ARM) performs VM mapping faster than copying,
586 then you should select this. This causes zsmalloc to use page table
587 mapping rather than copying for object mapping.
588
589 You can check speed with zsmalloc benchmark:
590 https://github.com/spartacus06/zsmapbench
591
592config ZSMALLOC_STAT
593 bool "Export zsmalloc statistics"
594 depends on ZSMALLOC
595 select DEBUG_FS
596 help
597 This option enables code in the zsmalloc to collect various
598 statistics about whats happening in zsmalloc and exports that
599 information to userspace via debugfs.
600 If unsure, say N.
601
602config GENERIC_EARLY_IOREMAP
603 bool
604
605config MAX_STACK_SIZE_MB
606 int "Maximum user stack size for 32-bit processes (MB)"
607 default 80
608 range 8 2048
609 depends on STACK_GROWSUP && (!64BIT || COMPAT)
610 help
611 This is the maximum stack size in Megabytes in the VM layout of 32-bit
612 user processes when the stack grows upwards (currently only on parisc
613 arch). The stack will be located at the highest memory address minus
614 the given value, unless the RLIMIT_STACK hard limit is changed to a
615 smaller value in which case that is used.
616
617 A sane initial value is 80 MB.
618
619config DEFERRED_STRUCT_PAGE_INIT
620 bool "Defer initialisation of struct pages to kthreads"
621 depends on SPARSEMEM
622 depends on !NEED_PER_CPU_KM
623 depends on 64BIT
624 help
625 Ordinarily all struct pages are initialised during early boot in a
626 single thread. On very large machines this can take a considerable
627 amount of time. If this option is set, large machines will bring up
628 a subset of memmap at boot and then initialise the rest in parallel
629 by starting one-off "pgdatinitX" kernel thread for each node X. This
630 has a potential performance impact on processes running early in the
631 lifetime of the system until these kthreads finish the
632 initialisation.
633
634config IDLE_PAGE_TRACKING
635 bool "Enable idle page tracking"
636 depends on SYSFS && MMU
637 select PAGE_EXTENSION if !64BIT
638 help
639 This feature allows to estimate the amount of user pages that have
640 not been touched during a given period of time. This information can
641 be useful to tune memory cgroup limits and/or for job placement
642 within a compute cluster.
643
644 See Documentation/admin-guide/mm/idle_page_tracking.rst for
645 more details.
646
647config ARCH_HAS_PTE_DEVMAP
648 bool
649
650config ZONE_DEVICE
651 bool "Device memory (pmem, HMM, etc...) hotplug support"
652 depends on MEMORY_HOTPLUG
653 depends on MEMORY_HOTREMOVE
654 depends on SPARSEMEM_VMEMMAP
655 depends on ARCH_HAS_PTE_DEVMAP
656 select XARRAY_MULTI
657
658 help
659 Device memory hotplug support allows for establishing pmem,
660 or other device driver discovered memory regions, in the
661 memmap. This allows pfn_to_page() lookups of otherwise
662 "device-physical" addresses which is needed for using a DAX
663 mapping in an O_DIRECT operation, among other things.
664
665 If FS_DAX is enabled, then say Y.
666
667config DEV_PAGEMAP_OPS
668 bool
669
670#
671# Helpers to mirror range of the CPU page tables of a process into device page
672# tables.
673#
674config HMM_MIRROR
675 bool
676 depends on MMU
677 depends on MMU_NOTIFIER
678
679config DEVICE_PRIVATE
680 bool "Unaddressable device memory (GPU memory, ...)"
681 depends on ZONE_DEVICE
682 select DEV_PAGEMAP_OPS
683
684 help
685 Allows creation of struct pages to represent unaddressable device
686 memory; i.e., memory that is only accessible from the device (or
687 group of devices). You likely also want to select HMM_MIRROR.
688
689config FRAME_VECTOR
690 bool
691
692config ARCH_USES_HIGH_VMA_FLAGS
693 bool
694config ARCH_HAS_PKEYS
695 bool
696
697config PERCPU_STATS
698 bool "Collect percpu memory statistics"
699 help
700 This feature collects and exposes statistics via debugfs. The
701 information includes global and per chunk statistics, which can
702 be used to help understand percpu memory usage.
703
704config GUP_BENCHMARK
705 bool "Enable infrastructure for get_user_pages_fast() benchmarking"
706 help
707 Provides /sys/kernel/debug/gup_benchmark that helps with testing
708 performance of get_user_pages_fast().
709
710 See tools/testing/selftests/vm/gup_benchmark.c
711
712config GUP_GET_PTE_LOW_HIGH
713 bool
714
715config READ_ONLY_THP_FOR_FS
716 bool "Read-only THP for filesystems (EXPERIMENTAL)"
717 depends on TRANSPARENT_HUGE_PAGECACHE && SHMEM
718
719 help
720 Allow khugepaged to put read-only file-backed pages in THP.
721
722 This is marked experimental because it is a new feature. Write
723 support of file THPs will be developed in the next few release
724 cycles.
725
726config ARCH_HAS_PTE_SPECIAL
727 bool
728
729#
730# Some architectures require a special hugepage directory format that is
731# required to support multiple hugepage sizes. For example a4fe3ce76
732# "powerpc/mm: Allow more flexible layouts for hugepage pagetables"
733# introduced it on powerpc. This allows for a more flexible hugepage
734# pagetable layouts.
735#
736config ARCH_HAS_HUGEPD
737 bool
738
739endmenu