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 config MEMORY_ISOLATION
146 # Only be set on architectures that have completely implemented memory hotplug
147 # feature. If you are not sure, don't touch it.
149 config HAVE_BOOTMEM_INFO_NODE
152 # eventually, we can have this option just 'select SPARSEMEM'
153 config MEMORY_HOTPLUG
154 bool "Allow for memory hot-add"
155 depends on SPARSEMEM || X86_64_ACPI_NUMA
156 depends on ARCH_ENABLE_MEMORY_HOTPLUG
158 config MEMORY_HOTPLUG_SPARSE
160 depends on SPARSEMEM && MEMORY_HOTPLUG
162 config MEMORY_HOTPLUG_DEFAULT_ONLINE
163 bool "Online the newly added memory blocks by default"
164 depends on MEMORY_HOTPLUG
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.
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.
177 config 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
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.
192 config SPLIT_PTLOCK_CPUS
194 default "999999" if !MMU
195 default "999999" if ARM && !CPU_CACHE_VIPT
196 default "999999" if PARISC && !PA20
199 config ARCH_ENABLE_SPLIT_PMD_PTLOCK
203 # support for memory balloon
204 config MEMORY_BALLOON
208 # support for memory balloon compaction
209 config BALLOON_COMPACTION
210 bool "Allow for balloon memory compaction/migration"
212 depends on COMPACTION && MEMORY_BALLOON
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.
223 # support for memory compaction
225 bool "Allow for memory compaction"
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
240 # support for free page reporting
241 config PAGE_REPORTING
242 bool "Free page reporting"
245 Free page reporting allows for the incremental acquisition of
246 free pages from the buddy allocator for the purpose of reporting
247 those pages to another entity, such as a hypervisor, so that the
248 memory can be freed within the host for other uses.
251 # support for page migration
254 bool "Page migration"
256 depends on (NUMA || ARCH_ENABLE_MEMORY_HOTREMOVE || COMPACTION || CMA) && MMU
258 Allows the migration of the physical location of pages of processes
259 while the virtual addresses are not changed. This is useful in
260 two situations. The first is on NUMA systems to put pages nearer
261 to the processors accessing. The second is when allocating huge
262 pages as migration can relocate pages to satisfy a huge page
263 allocation instead of reclaiming.
265 config ARCH_ENABLE_HUGEPAGE_MIGRATION
268 config ARCH_ENABLE_THP_MIGRATION
272 def_bool (MEMORY_ISOLATION && COMPACTION) || CMA
274 config PHYS_ADDR_T_64BIT
278 bool "Enable bounce buffers"
280 depends on BLOCK && MMU && (ZONE_DMA || HIGHMEM)
282 Enable bounce buffers for devices that cannot access
283 the full range of memory available to the CPU. Enabled
284 by default when ZONE_DMA or HIGHMEM is selected, but you
285 may say n to override this.
290 An architecture should select this if it implements the
291 deprecated interface virt_to_bus(). All new architectures
292 should probably not select this.
301 bool "Enable KSM for page merging"
305 Enable Kernel Samepage Merging: KSM periodically scans those areas
306 of an application's address space that an app has advised may be
307 mergeable. When it finds pages of identical content, it replaces
308 the many instances by a single page with that content, so
309 saving memory until one or another app needs to modify the content.
310 Recommended for use with KVM, or with other duplicative applications.
311 See Documentation/vm/ksm.rst for more information: KSM is inactive
312 until a program has madvised that an area is MADV_MERGEABLE, and
313 root has set /sys/kernel/mm/ksm/run to 1 (if CONFIG_SYSFS is set).
315 config DEFAULT_MMAP_MIN_ADDR
316 int "Low address space to protect from user allocation"
320 This is the portion of low virtual memory which should be protected
321 from userspace allocation. Keeping a user from writing to low pages
322 can help reduce the impact of kernel NULL pointer bugs.
324 For most ia64, ppc64 and x86 users with lots of address space
325 a value of 65536 is reasonable and should cause no problems.
326 On arm and other archs it should not be higher than 32768.
327 Programs which use vm86 functionality or have some need to map
328 this low address space will need CAP_SYS_RAWIO or disable this
329 protection by setting the value to 0.
331 This value can be changed after boot using the
332 /proc/sys/vm/mmap_min_addr tunable.
334 config ARCH_SUPPORTS_MEMORY_FAILURE
337 config MEMORY_FAILURE
339 depends on ARCH_SUPPORTS_MEMORY_FAILURE
340 bool "Enable recovery from hardware memory errors"
341 select MEMORY_ISOLATION
344 Enables code to recover from some memory failures on systems
345 with MCA recovery. This allows a system to continue running
346 even when some of its memory has uncorrected errors. This requires
347 special hardware support and typically ECC memory.
349 config HWPOISON_INJECT
350 tristate "HWPoison pages injector"
351 depends on MEMORY_FAILURE && DEBUG_KERNEL && PROC_FS
352 select PROC_PAGE_MONITOR
354 config NOMMU_INITIAL_TRIM_EXCESS
355 int "Turn on mmap() excess space trimming before booting"
359 The NOMMU mmap() frequently needs to allocate large contiguous chunks
360 of memory on which to store mappings, but it can only ask the system
361 allocator for chunks in 2^N*PAGE_SIZE amounts - which is frequently
362 more than it requires. To deal with this, mmap() is able to trim off
363 the excess and return it to the allocator.
365 If trimming is enabled, the excess is trimmed off and returned to the
366 system allocator, which can cause extra fragmentation, particularly
367 if there are a lot of transient processes.
369 If trimming is disabled, the excess is kept, but not used, which for
370 long-term mappings means that the space is wasted.
372 Trimming can be dynamically controlled through a sysctl option
373 (/proc/sys/vm/nr_trim_pages) which specifies the minimum number of
374 excess pages there must be before trimming should occur, or zero if
375 no trimming is to occur.
377 This option specifies the initial value of this option. The default
378 of 1 says that all excess pages should be trimmed.
380 See Documentation/nommu-mmap.txt for more information.
382 config TRANSPARENT_HUGEPAGE
383 bool "Transparent Hugepage Support"
384 depends on HAVE_ARCH_TRANSPARENT_HUGEPAGE
388 Transparent Hugepages allows the kernel to use huge pages and
389 huge tlb transparently to the applications whenever possible.
390 This feature can improve computing performance to certain
391 applications by speeding up page faults during memory
392 allocation, by reducing the number of tlb misses and by speeding
393 up the pagetable walking.
395 If memory constrained on embedded, you may want to say N.
398 prompt "Transparent Hugepage Support sysfs defaults"
399 depends on TRANSPARENT_HUGEPAGE
400 default TRANSPARENT_HUGEPAGE_ALWAYS
402 Selects the sysfs defaults for Transparent Hugepage Support.
404 config TRANSPARENT_HUGEPAGE_ALWAYS
407 Enabling Transparent Hugepage always, can increase the
408 memory footprint of applications without a guaranteed
409 benefit but it will work automatically for all applications.
411 config TRANSPARENT_HUGEPAGE_MADVISE
414 Enabling Transparent Hugepage madvise, will only provide a
415 performance improvement benefit to the applications using
416 madvise(MADV_HUGEPAGE) but it won't risk to increase the
417 memory footprint of applications without a guaranteed
421 config ARCH_WANTS_THP_SWAP
426 depends on TRANSPARENT_HUGEPAGE && ARCH_WANTS_THP_SWAP && SWAP
428 Swap transparent huge pages in one piece, without splitting.
429 XXX: For now, swap cluster backing transparent huge page
430 will be split after swapout.
432 For selection by architectures with reasonable THP sizes.
435 # UP and nommu archs use km based percpu allocator
437 config NEED_PER_CPU_KM
443 bool "Enable cleancache driver to cache clean pages if tmem is present"
445 Cleancache can be thought of as a page-granularity victim cache
446 for clean pages that the kernel's pageframe replacement algorithm
447 (PFRA) would like to keep around, but can't since there isn't enough
448 memory. So when the PFRA "evicts" a page, it first attempts to use
449 cleancache code to put the data contained in that page into
450 "transcendent memory", memory that is not directly accessible or
451 addressable by the kernel and is of unknown and possibly
452 time-varying size. And when a cleancache-enabled
453 filesystem wishes to access a page in a file on disk, it first
454 checks cleancache to see if it already contains it; if it does,
455 the page is copied into the kernel and a disk access is avoided.
456 When a transcendent memory driver is available (such as zcache or
457 Xen transcendent memory), a significant I/O reduction
458 may be achieved. When none is available, all cleancache calls
459 are reduced to a single pointer-compare-against-NULL resulting
460 in a negligible performance hit.
462 If unsure, say Y to enable cleancache
465 bool "Enable frontswap to cache swap pages if tmem is present"
468 Frontswap is so named because it can be thought of as the opposite
469 of a "backing" store for a swap device. The data is stored into
470 "transcendent memory", memory that is not directly accessible or
471 addressable by the kernel and is of unknown and possibly
472 time-varying size. When space in transcendent memory is available,
473 a significant swap I/O reduction may be achieved. When none is
474 available, all frontswap calls are reduced to a single pointer-
475 compare-against-NULL resulting in a negligible performance hit
476 and swap data is stored as normal on the matching swap device.
478 If unsure, say Y to enable frontswap.
481 bool "Contiguous Memory Allocator"
484 select MEMORY_ISOLATION
486 This enables the Contiguous Memory Allocator which allows other
487 subsystems to allocate big physically-contiguous blocks of memory.
488 CMA reserves a region of memory and allows only movable pages to
489 be allocated from it. This way, the kernel can use the memory for
490 pagecache and when a subsystem requests for contiguous area, the
491 allocated pages are migrated away to serve the contiguous request.
496 bool "CMA debug messages (DEVELOPMENT)"
497 depends on DEBUG_KERNEL && CMA
499 Turns on debug messages in CMA. This produces KERN_DEBUG
500 messages for every CMA call as well as various messages while
501 processing calls such as dma_alloc_from_contiguous().
502 This option does not affect warning and error messages.
505 bool "CMA debugfs interface"
506 depends on CMA && DEBUG_FS
508 Turns on the DebugFS interface for CMA.
511 int "Maximum count of the CMA areas"
515 CMA allows to create CMA areas for particular purpose, mainly,
516 used as device private area. This parameter sets the maximum
517 number of CMA area in the system.
519 If unsure, leave the default value "7".
521 config MEM_SOFT_DIRTY
522 bool "Track memory changes"
523 depends on CHECKPOINT_RESTORE && HAVE_ARCH_SOFT_DIRTY && PROC_FS
524 select PROC_PAGE_MONITOR
526 This option enables memory changes tracking by introducing a
527 soft-dirty bit on pte-s. This bit it set when someone writes
528 into a page just as regular dirty bit, but unlike the latter
529 it can be cleared by hands.
531 See Documentation/admin-guide/mm/soft-dirty.rst for more details.
534 bool "Compressed cache for swap pages (EXPERIMENTAL)"
535 depends on FRONTSWAP && CRYPTO=y
538 A lightweight compressed cache for swap pages. It takes
539 pages that are in the process of being swapped out and attempts to
540 compress them into a dynamically allocated RAM-based memory pool.
541 This can result in a significant I/O reduction on swap device and,
542 in the case where decompressing from RAM is faster that swap device
543 reads, can also improve workload performance.
545 This is marked experimental because it is a new feature (as of
546 v3.11) that interacts heavily with memory reclaim. While these
547 interactions don't cause any known issues on simple memory setups,
548 they have not be fully explored on the large set of potential
549 configurations and workloads that exist.
552 prompt "Compressed cache for swap pages default compressor"
554 default ZSWAP_COMPRESSOR_DEFAULT_LZO
556 Selects the default compression algorithm for the compressed cache
559 For an overview what kind of performance can be expected from
560 a particular compression algorithm please refer to the benchmarks
561 available at the following LWN page:
562 https://lwn.net/Articles/751795/
564 If in doubt, select 'LZO'.
566 The selection made here can be overridden by using the kernel
567 command line 'zswap.compressor=' option.
569 config ZSWAP_COMPRESSOR_DEFAULT_DEFLATE
571 select CRYPTO_DEFLATE
573 Use the Deflate algorithm as the default compression algorithm.
575 config ZSWAP_COMPRESSOR_DEFAULT_LZO
579 Use the LZO algorithm as the default compression algorithm.
581 config ZSWAP_COMPRESSOR_DEFAULT_842
585 Use the 842 algorithm as the default compression algorithm.
587 config ZSWAP_COMPRESSOR_DEFAULT_LZ4
591 Use the LZ4 algorithm as the default compression algorithm.
593 config ZSWAP_COMPRESSOR_DEFAULT_LZ4HC
597 Use the LZ4HC algorithm as the default compression algorithm.
599 config ZSWAP_COMPRESSOR_DEFAULT_ZSTD
603 Use the zstd algorithm as the default compression algorithm.
606 config ZSWAP_COMPRESSOR_DEFAULT
609 default "deflate" if ZSWAP_COMPRESSOR_DEFAULT_DEFLATE
610 default "lzo" if ZSWAP_COMPRESSOR_DEFAULT_LZO
611 default "842" if ZSWAP_COMPRESSOR_DEFAULT_842
612 default "lz4" if ZSWAP_COMPRESSOR_DEFAULT_LZ4
613 default "lz4hc" if ZSWAP_COMPRESSOR_DEFAULT_LZ4HC
614 default "zstd" if ZSWAP_COMPRESSOR_DEFAULT_ZSTD
618 prompt "Compressed cache for swap pages default allocator"
620 default ZSWAP_ZPOOL_DEFAULT_ZBUD
622 Selects the default allocator for the compressed cache for
624 The default is 'zbud' for compatibility, however please do
625 read the description of each of the allocators below before
626 making a right choice.
628 The selection made here can be overridden by using the kernel
629 command line 'zswap.zpool=' option.
631 config ZSWAP_ZPOOL_DEFAULT_ZBUD
635 Use the zbud allocator as the default allocator.
637 config ZSWAP_ZPOOL_DEFAULT_Z3FOLD
641 Use the z3fold allocator as the default allocator.
643 config ZSWAP_ZPOOL_DEFAULT_ZSMALLOC
647 Use the zsmalloc allocator as the default allocator.
650 config ZSWAP_ZPOOL_DEFAULT
653 default "zbud" if ZSWAP_ZPOOL_DEFAULT_ZBUD
654 default "z3fold" if ZSWAP_ZPOOL_DEFAULT_Z3FOLD
655 default "zsmalloc" if ZSWAP_ZPOOL_DEFAULT_ZSMALLOC
658 config ZSWAP_DEFAULT_ON
659 bool "Enable the compressed cache for swap pages by default"
662 If selected, the compressed cache for swap pages will be enabled
663 at boot, otherwise it will be disabled.
665 The selection made here can be overridden by using the kernel
666 command line 'zswap.enabled=' option.
669 tristate "Common API for compressed memory storage"
671 Compressed memory storage API. This allows using either zbud or
675 tristate "Low (Up to 2x) density storage for compressed pages"
677 A special purpose allocator for storing compressed pages.
678 It is designed to store up to two compressed pages per physical
679 page. While this design limits storage density, it has simple and
680 deterministic reclaim properties that make it preferable to a higher
681 density approach when reclaim will be used.
684 tristate "Up to 3x density storage for compressed pages"
687 A special purpose allocator for storing compressed pages.
688 It is designed to store up to three compressed pages per physical
689 page. It is a ZBUD derivative so the simplicity and determinism are
693 tristate "Memory allocator for compressed pages"
696 zsmalloc is a slab-based memory allocator designed to store
697 compressed RAM pages. zsmalloc uses virtual memory mapping
698 in order to reduce fragmentation. However, this results in a
699 non-standard allocator interface where a handle, not a pointer, is
700 returned by an alloc(). This handle must be mapped in order to
701 access the allocated space.
703 config PGTABLE_MAPPING
704 bool "Use page table mapping to access object in zsmalloc"
707 By default, zsmalloc uses a copy-based object mapping method to
708 access allocations that span two pages. However, if a particular
709 architecture (ex, ARM) performs VM mapping faster than copying,
710 then you should select this. This causes zsmalloc to use page table
711 mapping rather than copying for object mapping.
713 You can check speed with zsmalloc benchmark:
714 https://github.com/spartacus06/zsmapbench
717 bool "Export zsmalloc statistics"
721 This option enables code in the zsmalloc to collect various
722 statistics about whats happening in zsmalloc and exports that
723 information to userspace via debugfs.
726 config GENERIC_EARLY_IOREMAP
729 config MAX_STACK_SIZE_MB
730 int "Maximum user stack size for 32-bit processes (MB)"
733 depends on STACK_GROWSUP && (!64BIT || COMPAT)
735 This is the maximum stack size in Megabytes in the VM layout of 32-bit
736 user processes when the stack grows upwards (currently only on parisc
737 arch). The stack will be located at the highest memory address minus
738 the given value, unless the RLIMIT_STACK hard limit is changed to a
739 smaller value in which case that is used.
741 A sane initial value is 80 MB.
743 config DEFERRED_STRUCT_PAGE_INIT
744 bool "Defer initialisation of struct pages to kthreads"
746 depends on !NEED_PER_CPU_KM
749 Ordinarily all struct pages are initialised during early boot in a
750 single thread. On very large machines this can take a considerable
751 amount of time. If this option is set, large machines will bring up
752 a subset of memmap at boot and then initialise the rest in parallel
753 by starting one-off "pgdatinitX" kernel thread for each node X. This
754 has a potential performance impact on processes running early in the
755 lifetime of the system until these kthreads finish the
758 config IDLE_PAGE_TRACKING
759 bool "Enable idle page tracking"
760 depends on SYSFS && MMU
761 select PAGE_EXTENSION if !64BIT
763 This feature allows to estimate the amount of user pages that have
764 not been touched during a given period of time. This information can
765 be useful to tune memory cgroup limits and/or for job placement
766 within a compute cluster.
768 See Documentation/admin-guide/mm/idle_page_tracking.rst for
771 config ARCH_HAS_PTE_DEVMAP
775 bool "Device memory (pmem, HMM, etc...) hotplug support"
776 depends on MEMORY_HOTPLUG
777 depends on MEMORY_HOTREMOVE
778 depends on SPARSEMEM_VMEMMAP
779 depends on ARCH_HAS_PTE_DEVMAP
783 Device memory hotplug support allows for establishing pmem,
784 or other device driver discovered memory regions, in the
785 memmap. This allows pfn_to_page() lookups of otherwise
786 "device-physical" addresses which is needed for using a DAX
787 mapping in an O_DIRECT operation, among other things.
789 If FS_DAX is enabled, then say Y.
791 config DEV_PAGEMAP_OPS
795 # Helpers to mirror range of the CPU page tables of a process into device page
802 config DEVICE_PRIVATE
803 bool "Unaddressable device memory (GPU memory, ...)"
804 depends on ZONE_DEVICE
805 select DEV_PAGEMAP_OPS
808 Allows creation of struct pages to represent unaddressable device
809 memory; i.e., memory that is only accessible from the device (or
810 group of devices). You likely also want to select HMM_MIRROR.
815 config ARCH_USES_HIGH_VMA_FLAGS
817 config ARCH_HAS_PKEYS
821 bool "Collect percpu memory statistics"
823 This feature collects and exposes statistics via debugfs. The
824 information includes global and per chunk statistics, which can
825 be used to help understand percpu memory usage.
828 bool "Enable infrastructure for get_user_pages_fast() benchmarking"
830 Provides /sys/kernel/debug/gup_benchmark that helps with testing
831 performance of get_user_pages_fast().
833 See tools/testing/selftests/vm/gup_benchmark.c
835 config GUP_GET_PTE_LOW_HIGH
838 config READ_ONLY_THP_FOR_FS
839 bool "Read-only THP for filesystems (EXPERIMENTAL)"
840 depends on TRANSPARENT_HUGEPAGE && SHMEM
843 Allow khugepaged to put read-only file-backed pages in THP.
845 This is marked experimental because it is a new feature. Write
846 support of file THPs will be developed in the next few release
849 config ARCH_HAS_PTE_SPECIAL
853 # Some architectures require a special hugepage directory format that is
854 # required to support multiple hugepage sizes. For example a4fe3ce76
855 # "powerpc/mm: Allow more flexible layouts for hugepage pagetables"
856 # introduced it on powerpc. This allows for a more flexible hugepage
859 config ARCH_HAS_HUGEPD
862 config MAPPING_DIRTY_HELPERS