1 # SPDX-License-Identifier: GPL-2.0-only
3 menu "Memory Management options"
5 config SELECT_MEMORY_MODEL
7 depends on ARCH_SELECT_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
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.
23 depends on !(ARCH_DISCONTIGMEM_ENABLE || ARCH_SPARSEMEM_ENABLE) || ARCH_FLATMEM_ENABLE
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.
30 For systems that have holes in their physical address
31 spaces and for features like NUMA and memory hotplug,
32 choose "Sparse Memory".
34 If unsure, choose this option (Flat Memory) over any other.
36 config DISCONTIGMEM_MANUAL
37 bool "Discontiguous Memory"
38 depends on ARCH_DISCONTIGMEM_ENABLE
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.
45 Although "Discontiguous Memory" is still used by several
46 architectures, it is considered deprecated in favor of
49 If unsure, choose "Sparse Memory" over this option.
51 config SPARSEMEM_MANUAL
53 depends on ARCH_SPARSEMEM_ENABLE
55 This will be the only option for some systems, including
56 memory hot-plug systems. This is normal.
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.
62 If unsure, choose "Flat Memory" over this option.
68 depends on (!SELECT_MEMORY_MODEL && ARCH_DISCONTIGMEM_ENABLE) || DISCONTIGMEM_MANUAL
72 depends on (!SELECT_MEMORY_MODEL && ARCH_SPARSEMEM_ENABLE) || SPARSEMEM_MANUAL
76 depends on (!DISCONTIGMEM && !SPARSEMEM) || FLATMEM_MANUAL
78 config FLAT_NODE_MEM_MAP
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.
87 config NEED_MULTIPLE_NODES
89 depends on DISCONTIGMEM || NUMA
91 config HAVE_MEMORY_PRESENT
93 depends on ARCH_HAVE_MEMORY_PRESENT || SPARSEMEM
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.
102 # This option will also potentially produce smaller runtime code
103 # with gcc 3.4 and later.
105 config SPARSEMEM_STATIC
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.
113 config SPARSEMEM_EXTREME
115 depends on SPARSEMEM && !SPARSEMEM_STATIC
117 config SPARSEMEM_VMEMMAP_ENABLE
120 config SPARSEMEM_VMEMMAP
121 bool "Sparse Memory virtual memmap"
122 depends on SPARSEMEM && SPARSEMEM_VMEMMAP_ENABLE
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.
129 config HAVE_MEMBLOCK_NODE_MAP
132 config HAVE_MEMBLOCK_PHYS_MAP
139 config ARCH_KEEP_MEMBLOCK
142 # Keep arch NUMA mapping infrastructure post-init.
143 config NUMA_KEEP_MEMINFO
146 config MEMORY_ISOLATION
150 # Only be set on architectures that have completely implemented memory hotplug
151 # feature. If you are not sure, don't touch it.
153 config HAVE_BOOTMEM_INFO_NODE
156 # eventually, we can have this option just 'select SPARSEMEM'
157 config MEMORY_HOTPLUG
158 bool "Allow for memory hot-add"
159 depends on SPARSEMEM || X86_64_ACPI_NUMA
160 depends on ARCH_ENABLE_MEMORY_HOTPLUG
161 select NUMA_KEEP_MEMINFO if NUMA
163 config MEMORY_HOTPLUG_SPARSE
165 depends on SPARSEMEM && MEMORY_HOTPLUG
167 config MEMORY_HOTPLUG_DEFAULT_ONLINE
168 bool "Online the newly added memory blocks by default"
169 depends on MEMORY_HOTPLUG
171 This option sets the default policy setting for memory hotplug
172 onlining policy (/sys/devices/system/memory/auto_online_blocks) which
173 determines what happens to newly added memory regions. Policy setting
174 can always be changed at runtime.
175 See Documentation/admin-guide/mm/memory-hotplug.rst for more information.
177 Say Y here if you want all hot-plugged memory blocks to appear in
178 'online' state by default.
179 Say N here if you want the default policy to keep all hot-plugged
180 memory blocks in 'offline' state.
182 config MEMORY_HOTREMOVE
183 bool "Allow for memory hot remove"
184 select MEMORY_ISOLATION
185 select HAVE_BOOTMEM_INFO_NODE if (X86_64 || PPC64)
186 depends on MEMORY_HOTPLUG && ARCH_ENABLE_MEMORY_HOTREMOVE
189 # Heavily threaded applications may benefit from splitting the mm-wide
190 # page_table_lock, so that faults on different parts of the user address
191 # space can be handled with less contention: split it at this NR_CPUS.
192 # Default to 4 for wider testing, though 8 might be more appropriate.
193 # ARM's adjust_pte (unused if VIPT) depends on mm-wide page_table_lock.
194 # PA-RISC 7xxx's spinlock_t would enlarge struct page from 32 to 44 bytes.
195 # DEBUG_SPINLOCK and DEBUG_LOCK_ALLOC spinlock_t also enlarge struct page.
197 config SPLIT_PTLOCK_CPUS
199 default "999999" if !MMU
200 default "999999" if ARM && !CPU_CACHE_VIPT
201 default "999999" if PARISC && !PA20
204 config ARCH_ENABLE_SPLIT_PMD_PTLOCK
208 # support for memory balloon
209 config MEMORY_BALLOON
213 # support for memory balloon compaction
214 config BALLOON_COMPACTION
215 bool "Allow for balloon memory compaction/migration"
217 depends on COMPACTION && MEMORY_BALLOON
219 Memory fragmentation introduced by ballooning might reduce
220 significantly the number of 2MB contiguous memory blocks that can be
221 used within a guest, thus imposing performance penalties associated
222 with the reduced number of transparent huge pages that could be used
223 by the guest workload. Allowing the compaction & migration for memory
224 pages enlisted as being part of memory balloon devices avoids the
225 scenario aforementioned and helps improving memory defragmentation.
228 # support for memory compaction
230 bool "Allow for memory compaction"
235 Compaction is the only memory management component to form
236 high order (larger physically contiguous) memory blocks
237 reliably. The page allocator relies on compaction heavily and
238 the lack of the feature can lead to unexpected OOM killer
239 invocations for high order memory requests. You shouldn't
240 disable this option unless there really is a strong reason for
241 it and then we would be really interested to hear about that at
245 # support for page migration
248 bool "Page migration"
250 depends on (NUMA || ARCH_ENABLE_MEMORY_HOTREMOVE || COMPACTION || CMA) && MMU
252 Allows the migration of the physical location of pages of processes
253 while the virtual addresses are not changed. This is useful in
254 two situations. The first is on NUMA systems to put pages nearer
255 to the processors accessing. The second is when allocating huge
256 pages as migration can relocate pages to satisfy a huge page
257 allocation instead of reclaiming.
259 config ARCH_ENABLE_HUGEPAGE_MIGRATION
262 config ARCH_ENABLE_THP_MIGRATION
266 def_bool (MEMORY_ISOLATION && COMPACTION) || CMA
268 config PHYS_ADDR_T_64BIT
272 bool "Enable bounce buffers"
274 depends on BLOCK && MMU && (ZONE_DMA || HIGHMEM)
276 Enable bounce buffers for devices that cannot access
277 the full range of memory available to the CPU. Enabled
278 by default when ZONE_DMA or HIGHMEM is selected, but you
279 may say n to override this.
284 An architecture should select this if it implements the
285 deprecated interface virt_to_bus(). All new architectures
286 should probably not select this.
295 bool "Enable KSM for page merging"
299 Enable Kernel Samepage Merging: KSM periodically scans those areas
300 of an application's address space that an app has advised may be
301 mergeable. When it finds pages of identical content, it replaces
302 the many instances by a single page with that content, so
303 saving memory until one or another app needs to modify the content.
304 Recommended for use with KVM, or with other duplicative applications.
305 See Documentation/vm/ksm.rst for more information: KSM is inactive
306 until a program has madvised that an area is MADV_MERGEABLE, and
307 root has set /sys/kernel/mm/ksm/run to 1 (if CONFIG_SYSFS is set).
309 config DEFAULT_MMAP_MIN_ADDR
310 int "Low address space to protect from user allocation"
314 This is the portion of low virtual memory which should be protected
315 from userspace allocation. Keeping a user from writing to low pages
316 can help reduce the impact of kernel NULL pointer bugs.
318 For most ia64, ppc64 and x86 users with lots of address space
319 a value of 65536 is reasonable and should cause no problems.
320 On arm and other archs it should not be higher than 32768.
321 Programs which use vm86 functionality or have some need to map
322 this low address space will need CAP_SYS_RAWIO or disable this
323 protection by setting the value to 0.
325 This value can be changed after boot using the
326 /proc/sys/vm/mmap_min_addr tunable.
328 config ARCH_SUPPORTS_MEMORY_FAILURE
331 config MEMORY_FAILURE
333 depends on ARCH_SUPPORTS_MEMORY_FAILURE
334 bool "Enable recovery from hardware memory errors"
335 select MEMORY_ISOLATION
338 Enables code to recover from some memory failures on systems
339 with MCA recovery. This allows a system to continue running
340 even when some of its memory has uncorrected errors. This requires
341 special hardware support and typically ECC memory.
343 config HWPOISON_INJECT
344 tristate "HWPoison pages injector"
345 depends on MEMORY_FAILURE && DEBUG_KERNEL && PROC_FS
346 select PROC_PAGE_MONITOR
348 config NOMMU_INITIAL_TRIM_EXCESS
349 int "Turn on mmap() excess space trimming before booting"
353 The NOMMU mmap() frequently needs to allocate large contiguous chunks
354 of memory on which to store mappings, but it can only ask the system
355 allocator for chunks in 2^N*PAGE_SIZE amounts - which is frequently
356 more than it requires. To deal with this, mmap() is able to trim off
357 the excess and return it to the allocator.
359 If trimming is enabled, the excess is trimmed off and returned to the
360 system allocator, which can cause extra fragmentation, particularly
361 if there are a lot of transient processes.
363 If trimming is disabled, the excess is kept, but not used, which for
364 long-term mappings means that the space is wasted.
366 Trimming can be dynamically controlled through a sysctl option
367 (/proc/sys/vm/nr_trim_pages) which specifies the minimum number of
368 excess pages there must be before trimming should occur, or zero if
369 no trimming is to occur.
371 This option specifies the initial value of this option. The default
372 of 1 says that all excess pages should be trimmed.
374 See Documentation/nommu-mmap.txt for more information.
376 config TRANSPARENT_HUGEPAGE
377 bool "Transparent Hugepage Support"
378 depends on HAVE_ARCH_TRANSPARENT_HUGEPAGE
382 Transparent Hugepages allows the kernel to use huge pages and
383 huge tlb transparently to the applications whenever possible.
384 This feature can improve computing performance to certain
385 applications by speeding up page faults during memory
386 allocation, by reducing the number of tlb misses and by speeding
387 up the pagetable walking.
389 If memory constrained on embedded, you may want to say N.
392 prompt "Transparent Hugepage Support sysfs defaults"
393 depends on TRANSPARENT_HUGEPAGE
394 default TRANSPARENT_HUGEPAGE_ALWAYS
396 Selects the sysfs defaults for Transparent Hugepage Support.
398 config TRANSPARENT_HUGEPAGE_ALWAYS
401 Enabling Transparent Hugepage always, can increase the
402 memory footprint of applications without a guaranteed
403 benefit but it will work automatically for all applications.
405 config TRANSPARENT_HUGEPAGE_MADVISE
408 Enabling Transparent Hugepage madvise, will only provide a
409 performance improvement benefit to the applications using
410 madvise(MADV_HUGEPAGE) but it won't risk to increase the
411 memory footprint of applications without a guaranteed
415 config ARCH_WANTS_THP_SWAP
420 depends on TRANSPARENT_HUGEPAGE && ARCH_WANTS_THP_SWAP && SWAP
422 Swap transparent huge pages in one piece, without splitting.
423 XXX: For now, swap cluster backing transparent huge page
424 will be split after swapout.
426 For selection by architectures with reasonable THP sizes.
428 config TRANSPARENT_HUGE_PAGECACHE
430 depends on TRANSPARENT_HUGEPAGE
433 # UP and nommu archs use km based percpu allocator
435 config NEED_PER_CPU_KM
441 bool "Enable cleancache driver to cache clean pages if tmem is present"
443 Cleancache can be thought of as a page-granularity victim cache
444 for clean pages that the kernel's pageframe replacement algorithm
445 (PFRA) would like to keep around, but can't since there isn't enough
446 memory. So when the PFRA "evicts" a page, it first attempts to use
447 cleancache code to put the data contained in that page into
448 "transcendent memory", memory that is not directly accessible or
449 addressable by the kernel and is of unknown and possibly
450 time-varying size. And when a cleancache-enabled
451 filesystem wishes to access a page in a file on disk, it first
452 checks cleancache to see if it already contains it; if it does,
453 the page is copied into the kernel and a disk access is avoided.
454 When a transcendent memory driver is available (such as zcache or
455 Xen transcendent memory), a significant I/O reduction
456 may be achieved. When none is available, all cleancache calls
457 are reduced to a single pointer-compare-against-NULL resulting
458 in a negligible performance hit.
460 If unsure, say Y to enable cleancache
463 bool "Enable frontswap to cache swap pages if tmem is present"
466 Frontswap is so named because it can be thought of as the opposite
467 of a "backing" store for a swap device. The data is stored into
468 "transcendent memory", memory that is not directly accessible or
469 addressable by the kernel and is of unknown and possibly
470 time-varying size. When space in transcendent memory is available,
471 a significant swap I/O reduction may be achieved. When none is
472 available, all frontswap calls are reduced to a single pointer-
473 compare-against-NULL resulting in a negligible performance hit
474 and swap data is stored as normal on the matching swap device.
476 If unsure, say Y to enable frontswap.
479 bool "Contiguous Memory Allocator"
482 select MEMORY_ISOLATION
484 This enables the Contiguous Memory Allocator which allows other
485 subsystems to allocate big physically-contiguous blocks of memory.
486 CMA reserves a region of memory and allows only movable pages to
487 be allocated from it. This way, the kernel can use the memory for
488 pagecache and when a subsystem requests for contiguous area, the
489 allocated pages are migrated away to serve the contiguous request.
494 bool "CMA debug messages (DEVELOPMENT)"
495 depends on DEBUG_KERNEL && CMA
497 Turns on debug messages in CMA. This produces KERN_DEBUG
498 messages for every CMA call as well as various messages while
499 processing calls such as dma_alloc_from_contiguous().
500 This option does not affect warning and error messages.
503 bool "CMA debugfs interface"
504 depends on CMA && DEBUG_FS
506 Turns on the DebugFS interface for CMA.
509 int "Maximum count of the CMA areas"
513 CMA allows to create CMA areas for particular purpose, mainly,
514 used as device private area. This parameter sets the maximum
515 number of CMA area in the system.
517 If unsure, leave the default value "7".
519 config MEM_SOFT_DIRTY
520 bool "Track memory changes"
521 depends on CHECKPOINT_RESTORE && HAVE_ARCH_SOFT_DIRTY && PROC_FS
522 select PROC_PAGE_MONITOR
524 This option enables memory changes tracking by introducing a
525 soft-dirty bit on pte-s. This bit it set when someone writes
526 into a page just as regular dirty bit, but unlike the latter
527 it can be cleared by hands.
529 See Documentation/admin-guide/mm/soft-dirty.rst for more details.
532 bool "Compressed cache for swap pages (EXPERIMENTAL)"
533 depends on FRONTSWAP && CRYPTO=y
537 A lightweight compressed cache for swap pages. It takes
538 pages that are in the process of being swapped out and attempts to
539 compress them into a dynamically allocated RAM-based memory pool.
540 This can result in a significant I/O reduction on swap device and,
541 in the case where decompressing from RAM is faster that swap device
542 reads, can also improve workload performance.
544 This is marked experimental because it is a new feature (as of
545 v3.11) that interacts heavily with memory reclaim. While these
546 interactions don't cause any known issues on simple memory setups,
547 they have not be fully explored on the large set of potential
548 configurations and workloads that exist.
551 tristate "Common API for compressed memory storage"
553 Compressed memory storage API. This allows using either zbud or
557 tristate "Low (Up to 2x) density storage for compressed pages"
559 A special purpose allocator for storing compressed pages.
560 It is designed to store up to two compressed pages per physical
561 page. While this design limits storage density, it has simple and
562 deterministic reclaim properties that make it preferable to a higher
563 density approach when reclaim will be used.
566 tristate "Up to 3x density storage for compressed pages"
569 A special purpose allocator for storing compressed pages.
570 It is designed to store up to three compressed pages per physical
571 page. It is a ZBUD derivative so the simplicity and determinism are
575 tristate "Memory allocator for compressed pages"
578 zsmalloc is a slab-based memory allocator designed to store
579 compressed RAM pages. zsmalloc uses virtual memory mapping
580 in order to reduce fragmentation. However, this results in a
581 non-standard allocator interface where a handle, not a pointer, is
582 returned by an alloc(). This handle must be mapped in order to
583 access the allocated space.
585 config PGTABLE_MAPPING
586 bool "Use page table mapping to access object in zsmalloc"
589 By default, zsmalloc uses a copy-based object mapping method to
590 access allocations that span two pages. However, if a particular
591 architecture (ex, ARM) performs VM mapping faster than copying,
592 then you should select this. This causes zsmalloc to use page table
593 mapping rather than copying for object mapping.
595 You can check speed with zsmalloc benchmark:
596 https://github.com/spartacus06/zsmapbench
599 bool "Export zsmalloc statistics"
603 This option enables code in the zsmalloc to collect various
604 statistics about whats happening in zsmalloc and exports that
605 information to userspace via debugfs.
608 config GENERIC_EARLY_IOREMAP
611 config MAX_STACK_SIZE_MB
612 int "Maximum user stack size for 32-bit processes (MB)"
615 depends on STACK_GROWSUP && (!64BIT || COMPAT)
617 This is the maximum stack size in Megabytes in the VM layout of 32-bit
618 user processes when the stack grows upwards (currently only on parisc
619 arch). The stack will be located at the highest memory address minus
620 the given value, unless the RLIMIT_STACK hard limit is changed to a
621 smaller value in which case that is used.
623 A sane initial value is 80 MB.
625 config DEFERRED_STRUCT_PAGE_INIT
626 bool "Defer initialisation of struct pages to kthreads"
628 depends on !NEED_PER_CPU_KM
631 Ordinarily all struct pages are initialised during early boot in a
632 single thread. On very large machines this can take a considerable
633 amount of time. If this option is set, large machines will bring up
634 a subset of memmap at boot and then initialise the rest in parallel
635 by starting one-off "pgdatinitX" kernel thread for each node X. This
636 has a potential performance impact on processes running early in the
637 lifetime of the system until these kthreads finish the
640 config IDLE_PAGE_TRACKING
641 bool "Enable idle page tracking"
642 depends on SYSFS && MMU
643 select PAGE_EXTENSION if !64BIT
645 This feature allows to estimate the amount of user pages that have
646 not been touched during a given period of time. This information can
647 be useful to tune memory cgroup limits and/or for job placement
648 within a compute cluster.
650 See Documentation/admin-guide/mm/idle_page_tracking.rst for
653 config ARCH_HAS_PTE_DEVMAP
657 bool "Device memory (pmem, HMM, etc...) hotplug support"
658 depends on MEMORY_HOTPLUG
659 depends on MEMORY_HOTREMOVE
660 depends on SPARSEMEM_VMEMMAP
661 depends on ARCH_HAS_PTE_DEVMAP
665 Device memory hotplug support allows for establishing pmem,
666 or other device driver discovered memory regions, in the
667 memmap. This allows pfn_to_page() lookups of otherwise
668 "device-physical" addresses which is needed for using a DAX
669 mapping in an O_DIRECT operation, among other things.
671 If FS_DAX is enabled, then say Y.
673 config DEV_PAGEMAP_OPS
677 # Helpers to mirror range of the CPU page tables of a process into device page
684 config DEVICE_PRIVATE
685 bool "Unaddressable device memory (GPU memory, ...)"
686 depends on ZONE_DEVICE
687 select DEV_PAGEMAP_OPS
690 Allows creation of struct pages to represent unaddressable device
691 memory; i.e., memory that is only accessible from the device (or
692 group of devices). You likely also want to select HMM_MIRROR.
697 config ARCH_USES_HIGH_VMA_FLAGS
699 config ARCH_HAS_PKEYS
703 bool "Collect percpu memory statistics"
705 This feature collects and exposes statistics via debugfs. The
706 information includes global and per chunk statistics, which can
707 be used to help understand percpu memory usage.
710 bool "Enable infrastructure for get_user_pages_fast() benchmarking"
712 Provides /sys/kernel/debug/gup_benchmark that helps with testing
713 performance of get_user_pages_fast().
715 See tools/testing/selftests/vm/gup_benchmark.c
717 config GUP_GET_PTE_LOW_HIGH
720 config READ_ONLY_THP_FOR_FS
721 bool "Read-only THP for filesystems (EXPERIMENTAL)"
722 depends on TRANSPARENT_HUGE_PAGECACHE && SHMEM
725 Allow khugepaged to put read-only file-backed pages in THP.
727 This is marked experimental because it is a new feature. Write
728 support of file THPs will be developed in the next few release
731 config ARCH_HAS_PTE_SPECIAL
735 # Some architectures require a special hugepage directory format that is
736 # required to support multiple hugepage sizes. For example a4fe3ce76
737 # "powerpc/mm: Allow more flexible layouts for hugepage pagetables"
738 # introduced it on powerpc. This allows for a more flexible hugepage
741 config ARCH_HAS_HUGEPD
744 config MAPPING_DIRTY_HELPERS