1 # SPDX-License-Identifier: GPL-2.0-only
3 menu "Memory Management options"
6 # For some reason microblaze and nios2 hard code SWAP=n. Hopefully we can
7 # add proper SWAP support to them, in which case this can be remove.
16 bool "Support for paging of anonymous memory (swap)"
17 depends on MMU && BLOCK && !ARCH_NO_SWAP
20 This option allows you to choose whether you want to have support
21 for so called swap devices or swap files in your kernel that are
22 used to provide more virtual memory than the actual RAM present
23 in your computer. If unsure say Y.
26 bool "Compressed cache for swap pages"
31 A lightweight compressed cache for swap pages. It takes
32 pages that are in the process of being swapped out and attempts to
33 compress them into a dynamically allocated RAM-based memory pool.
34 This can result in a significant I/O reduction on swap device and,
35 in the case where decompressing from RAM is faster than swap device
36 reads, can also improve workload performance.
38 config ZSWAP_DEFAULT_ON
39 bool "Enable the compressed cache for swap pages by default"
42 If selected, the compressed cache for swap pages will be enabled
43 at boot, otherwise it will be disabled.
45 The selection made here can be overridden by using the kernel
46 command line 'zswap.enabled=' option.
48 config ZSWAP_EXCLUSIVE_LOADS_DEFAULT_ON
49 bool "Invalidate zswap entries when pages are loaded"
52 If selected, exclusive loads for zswap will be enabled at boot,
53 otherwise it will be disabled.
55 If exclusive loads are enabled, when a page is loaded from zswap,
56 the zswap entry is invalidated at once, as opposed to leaving it
57 in zswap until the swap entry is freed.
59 This avoids having two copies of the same page in memory
60 (compressed and uncompressed) after faulting in a page from zswap.
61 The cost is that if the page was never dirtied and needs to be
62 swapped out again, it will be re-compressed.
65 prompt "Default compressor"
67 default ZSWAP_COMPRESSOR_DEFAULT_LZO
69 Selects the default compression algorithm for the compressed cache
72 For an overview what kind of performance can be expected from
73 a particular compression algorithm please refer to the benchmarks
74 available at the following LWN page:
75 https://lwn.net/Articles/751795/
77 If in doubt, select 'LZO'.
79 The selection made here can be overridden by using the kernel
80 command line 'zswap.compressor=' option.
82 config ZSWAP_COMPRESSOR_DEFAULT_DEFLATE
86 Use the Deflate algorithm as the default compression algorithm.
88 config ZSWAP_COMPRESSOR_DEFAULT_LZO
92 Use the LZO algorithm as the default compression algorithm.
94 config ZSWAP_COMPRESSOR_DEFAULT_842
98 Use the 842 algorithm as the default compression algorithm.
100 config ZSWAP_COMPRESSOR_DEFAULT_LZ4
104 Use the LZ4 algorithm as the default compression algorithm.
106 config ZSWAP_COMPRESSOR_DEFAULT_LZ4HC
110 Use the LZ4HC algorithm as the default compression algorithm.
112 config ZSWAP_COMPRESSOR_DEFAULT_ZSTD
116 Use the zstd algorithm as the default compression algorithm.
119 config ZSWAP_COMPRESSOR_DEFAULT
122 default "deflate" if ZSWAP_COMPRESSOR_DEFAULT_DEFLATE
123 default "lzo" if ZSWAP_COMPRESSOR_DEFAULT_LZO
124 default "842" if ZSWAP_COMPRESSOR_DEFAULT_842
125 default "lz4" if ZSWAP_COMPRESSOR_DEFAULT_LZ4
126 default "lz4hc" if ZSWAP_COMPRESSOR_DEFAULT_LZ4HC
127 default "zstd" if ZSWAP_COMPRESSOR_DEFAULT_ZSTD
131 prompt "Default allocator"
133 default ZSWAP_ZPOOL_DEFAULT_ZBUD
135 Selects the default allocator for the compressed cache for
137 The default is 'zbud' for compatibility, however please do
138 read the description of each of the allocators below before
139 making a right choice.
141 The selection made here can be overridden by using the kernel
142 command line 'zswap.zpool=' option.
144 config ZSWAP_ZPOOL_DEFAULT_ZBUD
148 Use the zbud allocator as the default allocator.
150 config ZSWAP_ZPOOL_DEFAULT_Z3FOLD
154 Use the z3fold allocator as the default allocator.
156 config ZSWAP_ZPOOL_DEFAULT_ZSMALLOC
160 Use the zsmalloc allocator as the default allocator.
163 config ZSWAP_ZPOOL_DEFAULT
166 default "zbud" if ZSWAP_ZPOOL_DEFAULT_ZBUD
167 default "z3fold" if ZSWAP_ZPOOL_DEFAULT_Z3FOLD
168 default "zsmalloc" if ZSWAP_ZPOOL_DEFAULT_ZSMALLOC
172 tristate "2:1 compression allocator (zbud)"
175 A special purpose allocator for storing compressed pages.
176 It is designed to store up to two compressed pages per physical
177 page. While this design limits storage density, it has simple and
178 deterministic reclaim properties that make it preferable to a higher
179 density approach when reclaim will be used.
182 tristate "3:1 compression allocator (z3fold)"
185 A special purpose allocator for storing compressed pages.
186 It is designed to store up to three compressed pages per physical
187 page. It is a ZBUD derivative so the simplicity and determinism are
192 prompt "N:1 compression allocator (zsmalloc)" if ZSWAP
195 zsmalloc is a slab-based memory allocator designed to store
196 pages of various compression levels efficiently. It achieves
197 the highest storage density with the least amount of fragmentation.
200 bool "Export zsmalloc statistics"
204 This option enables code in the zsmalloc to collect various
205 statistics about what's happening in zsmalloc and exports that
206 information to userspace via debugfs.
209 config ZSMALLOC_CHAIN_SIZE
210 int "Maximum number of physical pages per-zspage"
215 This option sets the upper limit on the number of physical pages
216 that a zmalloc page (zspage) can consist of. The optimal zspage
217 chain size is calculated for each size class during the
218 initialization of the pool.
220 Changing this option can alter the characteristics of size classes,
221 such as the number of pages per zspage and the number of objects
222 per zspage. This can also result in different configurations of
223 the pool, as zsmalloc merges size classes with similar
226 For more information, see zsmalloc documentation.
228 menu "SLAB allocator options"
231 prompt "Choose SLAB allocator"
234 This option allows to select a slab allocator.
236 config SLAB_DEPRECATED
237 bool "SLAB (DEPRECATED)"
238 depends on !PREEMPT_RT
240 Deprecated and scheduled for removal in a few cycles. Replaced by
243 If you cannot migrate to SLUB, please contact linux-mm@kvack.org
244 and the people listed in the SLAB ALLOCATOR section of MAINTAINERS
245 file, explaining why.
247 The regular slab allocator that is established and known to work
248 well in all environments. It organizes cache hot objects in
249 per cpu and per node queues.
252 bool "SLUB (Unqueued Allocator)"
254 SLUB is a slab allocator that minimizes cache line usage
255 instead of managing queues of cached objects (SLAB approach).
256 Per cpu caching is realized using slabs of objects instead
257 of queues of objects. SLUB can use memory efficiently
258 and has enhanced diagnostics. SLUB is the default choice for
266 depends on SLAB_DEPRECATED
269 bool "Configure SLUB for minimal memory footprint"
270 depends on SLUB && EXPERT
271 select SLAB_MERGE_DEFAULT
273 Configures the SLUB allocator in a way to achieve minimal memory
274 footprint, sacrificing scalability, debugging and other features.
275 This is intended only for the smallest system that had used the
276 SLOB allocator and is not recommended for systems with more than
281 config SLAB_MERGE_DEFAULT
282 bool "Allow slab caches to be merged"
284 depends on SLAB || SLUB
286 For reduced kernel memory fragmentation, slab caches can be
287 merged when they share the same size and other characteristics.
288 This carries a risk of kernel heap overflows being able to
289 overwrite objects from merged caches (and more easily control
290 cache layout), which makes such heap attacks easier to exploit
291 by attackers. By keeping caches unmerged, these kinds of exploits
292 can usually only damage objects in the same cache. To disable
293 merging at runtime, "slab_nomerge" can be passed on the kernel
296 config SLAB_FREELIST_RANDOM
297 bool "Randomize slab freelist"
298 depends on SLAB || (SLUB && !SLUB_TINY)
300 Randomizes the freelist order used on creating new pages. This
301 security feature reduces the predictability of the kernel slab
302 allocator against heap overflows.
304 config SLAB_FREELIST_HARDENED
305 bool "Harden slab freelist metadata"
306 depends on SLAB || (SLUB && !SLUB_TINY)
308 Many kernel heap attacks try to target slab cache metadata and
309 other infrastructure. This options makes minor performance
310 sacrifices to harden the kernel slab allocator against common
311 freelist exploit methods. Some slab implementations have more
312 sanity-checking than others. This option is most effective with
317 bool "Enable SLUB performance statistics"
318 depends on SLUB && SYSFS && !SLUB_TINY
320 SLUB statistics are useful to debug SLUBs allocation behavior in
321 order find ways to optimize the allocator. This should never be
322 enabled for production use since keeping statistics slows down
323 the allocator by a few percentage points. The slabinfo command
324 supports the determination of the most active slabs to figure
325 out which slabs are relevant to a particular load.
326 Try running: slabinfo -DA
328 config SLUB_CPU_PARTIAL
330 depends on SLUB && SMP && !SLUB_TINY
331 bool "SLUB per cpu partial cache"
333 Per cpu partial caches accelerate objects allocation and freeing
334 that is local to a processor at the price of more indeterminism
335 in the latency of the free. On overflow these caches will be cleared
336 which requires the taking of locks that may cause latency spikes.
337 Typically one would choose no for a realtime system.
339 endmenu # SLAB allocator options
341 config SHUFFLE_PAGE_ALLOCATOR
342 bool "Page allocator randomization"
343 default SLAB_FREELIST_RANDOM && ACPI_NUMA
345 Randomization of the page allocator improves the average
346 utilization of a direct-mapped memory-side-cache. See section
347 5.2.27 Heterogeneous Memory Attribute Table (HMAT) in the ACPI
348 6.2a specification for an example of how a platform advertises
349 the presence of a memory-side-cache. There are also incidental
350 security benefits as it reduces the predictability of page
351 allocations to compliment SLAB_FREELIST_RANDOM, but the
352 default granularity of shuffling on the MAX_ORDER i.e, 10th
353 order of pages is selected based on cache utilization benefits
356 While the randomization improves cache utilization it may
357 negatively impact workloads on platforms without a cache. For
358 this reason, by default, the randomization is enabled only
359 after runtime detection of a direct-mapped memory-side-cache.
360 Otherwise, the randomization may be force enabled with the
361 'page_alloc.shuffle' kernel command line parameter.
366 bool "Disable heap randomization"
369 Randomizing heap placement makes heap exploits harder, but it
370 also breaks ancient binaries (including anything libc5 based).
371 This option changes the bootup default to heap randomization
372 disabled, and can be overridden at runtime by setting
373 /proc/sys/kernel/randomize_va_space to 2.
375 On non-ancient distros (post-2000 ones) N is usually a safe choice.
377 config MMAP_ALLOW_UNINITIALIZED
378 bool "Allow mmapped anonymous memory to be uninitialized"
379 depends on EXPERT && !MMU
382 Normally, and according to the Linux spec, anonymous memory obtained
383 from mmap() has its contents cleared before it is passed to
384 userspace. Enabling this config option allows you to request that
385 mmap() skip that if it is given an MAP_UNINITIALIZED flag, thus
386 providing a huge performance boost. If this option is not enabled,
387 then the flag will be ignored.
389 This is taken advantage of by uClibc's malloc(), and also by
390 ELF-FDPIC binfmt's brk and stack allocator.
392 Because of the obvious security issues, this option should only be
393 enabled on embedded devices where you control what is run in
394 userspace. Since that isn't generally a problem on no-MMU systems,
395 it is normally safe to say Y here.
397 See Documentation/admin-guide/mm/nommu-mmap.rst for more information.
399 config SELECT_MEMORY_MODEL
401 depends on ARCH_SELECT_MEMORY_MODEL
404 prompt "Memory model"
405 depends on SELECT_MEMORY_MODEL
406 default SPARSEMEM_MANUAL if ARCH_SPARSEMEM_DEFAULT
407 default FLATMEM_MANUAL
409 This option allows you to change some of the ways that
410 Linux manages its memory internally. Most users will
411 only have one option here selected by the architecture
412 configuration. This is normal.
414 config FLATMEM_MANUAL
416 depends on !ARCH_SPARSEMEM_ENABLE || ARCH_FLATMEM_ENABLE
418 This option is best suited for non-NUMA systems with
419 flat address space. The FLATMEM is the most efficient
420 system in terms of performance and resource consumption
421 and it is the best option for smaller systems.
423 For systems that have holes in their physical address
424 spaces and for features like NUMA and memory hotplug,
425 choose "Sparse Memory".
427 If unsure, choose this option (Flat Memory) over any other.
429 config SPARSEMEM_MANUAL
431 depends on ARCH_SPARSEMEM_ENABLE
433 This will be the only option for some systems, including
434 memory hot-plug systems. This is normal.
436 This option provides efficient support for systems with
437 holes is their physical address space and allows memory
438 hot-plug and hot-remove.
440 If unsure, choose "Flat Memory" over this option.
446 depends on (!SELECT_MEMORY_MODEL && ARCH_SPARSEMEM_ENABLE) || SPARSEMEM_MANUAL
450 depends on !SPARSEMEM || FLATMEM_MANUAL
453 # SPARSEMEM_EXTREME (which is the default) does some bootmem
454 # allocations when sparse_init() is called. If this cannot
455 # be done on your architecture, select this option. However,
456 # statically allocating the mem_section[] array can potentially
457 # consume vast quantities of .bss, so be careful.
459 # This option will also potentially produce smaller runtime code
460 # with gcc 3.4 and later.
462 config SPARSEMEM_STATIC
466 # Architecture platforms which require a two level mem_section in SPARSEMEM
467 # must select this option. This is usually for architecture platforms with
468 # an extremely sparse physical address space.
470 config SPARSEMEM_EXTREME
472 depends on SPARSEMEM && !SPARSEMEM_STATIC
474 config SPARSEMEM_VMEMMAP_ENABLE
477 config SPARSEMEM_VMEMMAP
478 bool "Sparse Memory virtual memmap"
479 depends on SPARSEMEM && SPARSEMEM_VMEMMAP_ENABLE
482 SPARSEMEM_VMEMMAP uses a virtually mapped memmap to optimise
483 pfn_to_page and page_to_pfn operations. This is the most
484 efficient option when sufficient kernel resources are available.
486 # Select this config option from the architecture Kconfig, if it is preferred
487 # to enable the feature of HugeTLB/dev_dax vmemmap optimization.
489 config ARCH_WANT_OPTIMIZE_DAX_VMEMMAP
492 config ARCH_WANT_OPTIMIZE_HUGETLB_VMEMMAP
495 config HAVE_MEMBLOCK_PHYS_MAP
502 # Don't discard allocated memory used to track "memory" and "reserved" memblocks
503 # after early boot, so it can still be used to test for validity of memory.
504 # Also, memblocks are updated with memory hot(un)plug.
505 config ARCH_KEEP_MEMBLOCK
508 # Keep arch NUMA mapping infrastructure post-init.
509 config NUMA_KEEP_MEMINFO
512 config MEMORY_ISOLATION
515 # IORESOURCE_SYSTEM_RAM regions in the kernel resource tree that are marked
516 # IORESOURCE_EXCLUSIVE cannot be mapped to user space, for example, via
518 config EXCLUSIVE_SYSTEM_RAM
520 depends on !DEVMEM || STRICT_DEVMEM
523 # Only be set on architectures that have completely implemented memory hotplug
524 # feature. If you are not sure, don't touch it.
526 config HAVE_BOOTMEM_INFO_NODE
529 config ARCH_ENABLE_MEMORY_HOTPLUG
532 config ARCH_ENABLE_MEMORY_HOTREMOVE
535 # eventually, we can have this option just 'select SPARSEMEM'
536 menuconfig MEMORY_HOTPLUG
537 bool "Memory hotplug"
538 select MEMORY_ISOLATION
540 depends on ARCH_ENABLE_MEMORY_HOTPLUG
542 select NUMA_KEEP_MEMINFO if NUMA
546 config MEMORY_HOTPLUG_DEFAULT_ONLINE
547 bool "Online the newly added memory blocks by default"
548 depends on MEMORY_HOTPLUG
550 This option sets the default policy setting for memory hotplug
551 onlining policy (/sys/devices/system/memory/auto_online_blocks) which
552 determines what happens to newly added memory regions. Policy setting
553 can always be changed at runtime.
554 See Documentation/admin-guide/mm/memory-hotplug.rst for more information.
556 Say Y here if you want all hot-plugged memory blocks to appear in
557 'online' state by default.
558 Say N here if you want the default policy to keep all hot-plugged
559 memory blocks in 'offline' state.
561 config MEMORY_HOTREMOVE
562 bool "Allow for memory hot remove"
563 select HAVE_BOOTMEM_INFO_NODE if (X86_64 || PPC64)
564 depends on MEMORY_HOTPLUG && ARCH_ENABLE_MEMORY_HOTREMOVE
567 config MHP_MEMMAP_ON_MEMORY
569 depends on MEMORY_HOTPLUG && SPARSEMEM_VMEMMAP
570 depends on ARCH_MHP_MEMMAP_ON_MEMORY_ENABLE
572 endif # MEMORY_HOTPLUG
574 config ARCH_MHP_MEMMAP_ON_MEMORY_ENABLE
577 # Heavily threaded applications may benefit from splitting the mm-wide
578 # page_table_lock, so that faults on different parts of the user address
579 # space can be handled with less contention: split it at this NR_CPUS.
580 # Default to 4 for wider testing, though 8 might be more appropriate.
581 # ARM's adjust_pte (unused if VIPT) depends on mm-wide page_table_lock.
582 # PA-RISC 7xxx's spinlock_t would enlarge struct page from 32 to 44 bytes.
583 # SPARC32 allocates multiple pte tables within a single page, and therefore
584 # a per-page lock leads to problems when multiple tables need to be locked
585 # at the same time (e.g. copy_page_range()).
586 # DEBUG_SPINLOCK and DEBUG_LOCK_ALLOC spinlock_t also enlarge struct page.
588 config SPLIT_PTLOCK_CPUS
590 default "999999" if !MMU
591 default "999999" if ARM && !CPU_CACHE_VIPT
592 default "999999" if PARISC && !PA20
593 default "999999" if SPARC32
596 config ARCH_ENABLE_SPLIT_PMD_PTLOCK
600 # support for memory balloon
601 config MEMORY_BALLOON
605 # support for memory balloon compaction
606 config BALLOON_COMPACTION
607 bool "Allow for balloon memory compaction/migration"
609 depends on COMPACTION && MEMORY_BALLOON
611 Memory fragmentation introduced by ballooning might reduce
612 significantly the number of 2MB contiguous memory blocks that can be
613 used within a guest, thus imposing performance penalties associated
614 with the reduced number of transparent huge pages that could be used
615 by the guest workload. Allowing the compaction & migration for memory
616 pages enlisted as being part of memory balloon devices avoids the
617 scenario aforementioned and helps improving memory defragmentation.
620 # support for memory compaction
622 bool "Allow for memory compaction"
627 Compaction is the only memory management component to form
628 high order (larger physically contiguous) memory blocks
629 reliably. The page allocator relies on compaction heavily and
630 the lack of the feature can lead to unexpected OOM killer
631 invocations for high order memory requests. You shouldn't
632 disable this option unless there really is a strong reason for
633 it and then we would be really interested to hear about that at
636 config COMPACT_UNEVICTABLE_DEFAULT
638 depends on COMPACTION
639 default 0 if PREEMPT_RT
643 # support for free page reporting
644 config PAGE_REPORTING
645 bool "Free page reporting"
648 Free page reporting allows for the incremental acquisition of
649 free pages from the buddy allocator for the purpose of reporting
650 those pages to another entity, such as a hypervisor, so that the
651 memory can be freed within the host for other uses.
654 # support for page migration
657 bool "Page migration"
659 depends on (NUMA || ARCH_ENABLE_MEMORY_HOTREMOVE || COMPACTION || CMA) && MMU
661 Allows the migration of the physical location of pages of processes
662 while the virtual addresses are not changed. This is useful in
663 two situations. The first is on NUMA systems to put pages nearer
664 to the processors accessing. The second is when allocating huge
665 pages as migration can relocate pages to satisfy a huge page
666 allocation instead of reclaiming.
668 config DEVICE_MIGRATION
669 def_bool MIGRATION && ZONE_DEVICE
671 config ARCH_ENABLE_HUGEPAGE_MIGRATION
674 config ARCH_ENABLE_THP_MIGRATION
677 config HUGETLB_PAGE_SIZE_VARIABLE
680 Allows the pageblock_order value to be dynamic instead of just standard
681 HUGETLB_PAGE_ORDER when there are multiple HugeTLB page sizes available
684 Note that the pageblock_order cannot exceed MAX_ORDER and will be
685 clamped down to MAX_ORDER.
688 def_bool (MEMORY_ISOLATION && COMPACTION) || CMA
690 config PHYS_ADDR_T_64BIT
694 bool "Enable bounce buffers"
696 depends on BLOCK && MMU && HIGHMEM
698 Enable bounce buffers for devices that cannot access the full range of
699 memory available to the CPU. Enabled by default when HIGHMEM is
700 selected, but you may say n to override this.
707 bool "Enable KSM for page merging"
711 Enable Kernel Samepage Merging: KSM periodically scans those areas
712 of an application's address space that an app has advised may be
713 mergeable. When it finds pages of identical content, it replaces
714 the many instances by a single page with that content, so
715 saving memory until one or another app needs to modify the content.
716 Recommended for use with KVM, or with other duplicative applications.
717 See Documentation/mm/ksm.rst for more information: KSM is inactive
718 until a program has madvised that an area is MADV_MERGEABLE, and
719 root has set /sys/kernel/mm/ksm/run to 1 (if CONFIG_SYSFS is set).
721 config DEFAULT_MMAP_MIN_ADDR
722 int "Low address space to protect from user allocation"
726 This is the portion of low virtual memory which should be protected
727 from userspace allocation. Keeping a user from writing to low pages
728 can help reduce the impact of kernel NULL pointer bugs.
730 For most ia64, ppc64 and x86 users with lots of address space
731 a value of 65536 is reasonable and should cause no problems.
732 On arm and other archs it should not be higher than 32768.
733 Programs which use vm86 functionality or have some need to map
734 this low address space will need CAP_SYS_RAWIO or disable this
735 protection by setting the value to 0.
737 This value can be changed after boot using the
738 /proc/sys/vm/mmap_min_addr tunable.
740 config ARCH_SUPPORTS_MEMORY_FAILURE
743 config MEMORY_FAILURE
745 depends on ARCH_SUPPORTS_MEMORY_FAILURE
746 bool "Enable recovery from hardware memory errors"
747 select MEMORY_ISOLATION
750 Enables code to recover from some memory failures on systems
751 with MCA recovery. This allows a system to continue running
752 even when some of its memory has uncorrected errors. This requires
753 special hardware support and typically ECC memory.
755 config HWPOISON_INJECT
756 tristate "HWPoison pages injector"
757 depends on MEMORY_FAILURE && DEBUG_KERNEL && PROC_FS
758 select PROC_PAGE_MONITOR
760 config NOMMU_INITIAL_TRIM_EXCESS
761 int "Turn on mmap() excess space trimming before booting"
765 The NOMMU mmap() frequently needs to allocate large contiguous chunks
766 of memory on which to store mappings, but it can only ask the system
767 allocator for chunks in 2^N*PAGE_SIZE amounts - which is frequently
768 more than it requires. To deal with this, mmap() is able to trim off
769 the excess and return it to the allocator.
771 If trimming is enabled, the excess is trimmed off and returned to the
772 system allocator, which can cause extra fragmentation, particularly
773 if there are a lot of transient processes.
775 If trimming is disabled, the excess is kept, but not used, which for
776 long-term mappings means that the space is wasted.
778 Trimming can be dynamically controlled through a sysctl option
779 (/proc/sys/vm/nr_trim_pages) which specifies the minimum number of
780 excess pages there must be before trimming should occur, or zero if
781 no trimming is to occur.
783 This option specifies the initial value of this option. The default
784 of 1 says that all excess pages should be trimmed.
786 See Documentation/admin-guide/mm/nommu-mmap.rst for more information.
788 config ARCH_WANT_GENERAL_HUGETLB
791 config ARCH_WANTS_THP_SWAP
794 menuconfig TRANSPARENT_HUGEPAGE
795 bool "Transparent Hugepage Support"
796 depends on HAVE_ARCH_TRANSPARENT_HUGEPAGE && !PREEMPT_RT
800 Transparent Hugepages allows the kernel to use huge pages and
801 huge tlb transparently to the applications whenever possible.
802 This feature can improve computing performance to certain
803 applications by speeding up page faults during memory
804 allocation, by reducing the number of tlb misses and by speeding
805 up the pagetable walking.
807 If memory constrained on embedded, you may want to say N.
809 if TRANSPARENT_HUGEPAGE
812 prompt "Transparent Hugepage Support sysfs defaults"
813 depends on TRANSPARENT_HUGEPAGE
814 default TRANSPARENT_HUGEPAGE_ALWAYS
816 Selects the sysfs defaults for Transparent Hugepage Support.
818 config TRANSPARENT_HUGEPAGE_ALWAYS
821 Enabling Transparent Hugepage always, can increase the
822 memory footprint of applications without a guaranteed
823 benefit but it will work automatically for all applications.
825 config TRANSPARENT_HUGEPAGE_MADVISE
828 Enabling Transparent Hugepage madvise, will only provide a
829 performance improvement benefit to the applications using
830 madvise(MADV_HUGEPAGE) but it won't risk to increase the
831 memory footprint of applications without a guaranteed
837 depends on TRANSPARENT_HUGEPAGE && ARCH_WANTS_THP_SWAP && SWAP && 64BIT
839 Swap transparent huge pages in one piece, without splitting.
840 XXX: For now, swap cluster backing transparent huge page
841 will be split after swapout.
843 For selection by architectures with reasonable THP sizes.
845 config READ_ONLY_THP_FOR_FS
846 bool "Read-only THP for filesystems (EXPERIMENTAL)"
847 depends on TRANSPARENT_HUGEPAGE && SHMEM
850 Allow khugepaged to put read-only file-backed pages in THP.
852 This is marked experimental because it is a new feature. Write
853 support of file THPs will be developed in the next few release
856 endif # TRANSPARENT_HUGEPAGE
859 # UP and nommu archs use km based percpu allocator
861 config NEED_PER_CPU_KM
862 depends on !SMP || !MMU
866 config NEED_PER_CPU_EMBED_FIRST_CHUNK
869 config NEED_PER_CPU_PAGE_FIRST_CHUNK
872 config USE_PERCPU_NUMA_NODE_ID
875 config HAVE_SETUP_PER_CPU_AREA
879 bool "Contiguous Memory Allocator"
882 select MEMORY_ISOLATION
884 This enables the Contiguous Memory Allocator which allows other
885 subsystems to allocate big physically-contiguous blocks of memory.
886 CMA reserves a region of memory and allows only movable pages to
887 be allocated from it. This way, the kernel can use the memory for
888 pagecache and when a subsystem requests for contiguous area, the
889 allocated pages are migrated away to serve the contiguous request.
894 bool "CMA debug messages (DEVELOPMENT)"
895 depends on DEBUG_KERNEL && CMA
897 Turns on debug messages in CMA. This produces KERN_DEBUG
898 messages for every CMA call as well as various messages while
899 processing calls such as dma_alloc_from_contiguous().
900 This option does not affect warning and error messages.
903 bool "CMA debugfs interface"
904 depends on CMA && DEBUG_FS
906 Turns on the DebugFS interface for CMA.
909 bool "CMA information through sysfs interface"
910 depends on CMA && SYSFS
912 This option exposes some sysfs attributes to get information
916 int "Maximum count of the CMA areas"
921 CMA allows to create CMA areas for particular purpose, mainly,
922 used as device private area. This parameter sets the maximum
923 number of CMA area in the system.
925 If unsure, leave the default value "7" in UMA and "19" in NUMA.
927 config MEM_SOFT_DIRTY
928 bool "Track memory changes"
929 depends on CHECKPOINT_RESTORE && HAVE_ARCH_SOFT_DIRTY && PROC_FS
930 select PROC_PAGE_MONITOR
932 This option enables memory changes tracking by introducing a
933 soft-dirty bit on pte-s. This bit it set when someone writes
934 into a page just as regular dirty bit, but unlike the latter
935 it can be cleared by hands.
937 See Documentation/admin-guide/mm/soft-dirty.rst for more details.
939 config GENERIC_EARLY_IOREMAP
942 config STACK_MAX_DEFAULT_SIZE_MB
943 int "Default maximum user stack size for 32-bit processes (MB)"
946 depends on STACK_GROWSUP && (!64BIT || COMPAT)
948 This is the maximum stack size in Megabytes in the VM layout of 32-bit
949 user processes when the stack grows upwards (currently only on parisc
950 arch) when the RLIMIT_STACK hard limit is unlimited.
952 A sane initial value is 100 MB.
954 config DEFERRED_STRUCT_PAGE_INIT
955 bool "Defer initialisation of struct pages to kthreads"
957 depends on !NEED_PER_CPU_KM
961 Ordinarily all struct pages are initialised during early boot in a
962 single thread. On very large machines this can take a considerable
963 amount of time. If this option is set, large machines will bring up
964 a subset of memmap at boot and then initialise the rest in parallel.
965 This has a potential performance impact on tasks running early in the
966 lifetime of the system until these kthreads finish the
969 config PAGE_IDLE_FLAG
971 select PAGE_EXTENSION if !64BIT
973 This adds PG_idle and PG_young flags to 'struct page'. PTE Accessed
974 bit writers can set the state of the bit in the flags so that PTE
975 Accessed bit readers may avoid disturbance.
977 config IDLE_PAGE_TRACKING
978 bool "Enable idle page tracking"
979 depends on SYSFS && MMU
980 select PAGE_IDLE_FLAG
982 This feature allows to estimate the amount of user pages that have
983 not been touched during a given period of time. This information can
984 be useful to tune memory cgroup limits and/or for job placement
985 within a compute cluster.
987 See Documentation/admin-guide/mm/idle_page_tracking.rst for
990 config ARCH_HAS_CACHE_LINE_SIZE
993 config ARCH_HAS_CURRENT_STACK_POINTER
996 In support of HARDENED_USERCOPY performing stack variable lifetime
997 checking, an architecture-agnostic way to find the stack pointer
998 is needed. Once an architecture defines an unsigned long global
999 register alias named "current_stack_pointer", this config can be
1002 config ARCH_HAS_PTE_DEVMAP
1005 config ARCH_HAS_ZONE_DMA_SET
1009 bool "Support DMA zone" if ARCH_HAS_ZONE_DMA_SET
1010 default y if ARM64 || X86
1013 bool "Support DMA32 zone" if ARCH_HAS_ZONE_DMA_SET
1018 bool "Device memory (pmem, HMM, etc...) hotplug support"
1019 depends on MEMORY_HOTPLUG
1020 depends on MEMORY_HOTREMOVE
1021 depends on SPARSEMEM_VMEMMAP
1022 depends on ARCH_HAS_PTE_DEVMAP
1026 Device memory hotplug support allows for establishing pmem,
1027 or other device driver discovered memory regions, in the
1028 memmap. This allows pfn_to_page() lookups of otherwise
1029 "device-physical" addresses which is needed for using a DAX
1030 mapping in an O_DIRECT operation, among other things.
1032 If FS_DAX is enabled, then say Y.
1035 # Helpers to mirror range of the CPU page tables of a process into device page
1042 config GET_FREE_REGION
1043 depends on SPARSEMEM
1046 config DEVICE_PRIVATE
1047 bool "Unaddressable device memory (GPU memory, ...)"
1048 depends on ZONE_DEVICE
1049 select GET_FREE_REGION
1052 Allows creation of struct pages to represent unaddressable device
1053 memory; i.e., memory that is only accessible from the device (or
1054 group of devices). You likely also want to select HMM_MIRROR.
1059 config ARCH_USES_HIGH_VMA_FLAGS
1061 config ARCH_HAS_PKEYS
1064 config ARCH_USES_PG_ARCH_X
1067 Enable the definition of PG_arch_x page flags with x > 1. Only
1068 suitable for 64-bit architectures with CONFIG_FLATMEM or
1069 CONFIG_SPARSEMEM_VMEMMAP enabled, otherwise there may not be
1070 enough room for additional bits in page->flags.
1072 config VM_EVENT_COUNTERS
1074 bool "Enable VM event counters for /proc/vmstat" if EXPERT
1076 VM event counters are needed for event counts to be shown.
1077 This option allows the disabling of the VM event counters
1078 on EXPERT systems. /proc/vmstat will only show page counts
1079 if VM event counters are disabled.
1082 bool "Collect percpu memory statistics"
1084 This feature collects and exposes statistics via debugfs. The
1085 information includes global and per chunk statistics, which can
1086 be used to help understand percpu memory usage.
1089 bool "Enable infrastructure for get_user_pages()-related unit tests"
1092 Provides /sys/kernel/debug/gup_test, which in turn provides a way
1093 to make ioctl calls that can launch kernel-based unit tests for
1094 the get_user_pages*() and pin_user_pages*() family of API calls.
1096 These tests include benchmark testing of the _fast variants of
1097 get_user_pages*() and pin_user_pages*(), as well as smoke tests of
1098 the non-_fast variants.
1100 There is also a sub-test that allows running dump_page() on any
1101 of up to eight pages (selected by command line args) within the
1102 range of user-space addresses. These pages are either pinned via
1103 pin_user_pages*(), or pinned via get_user_pages*(), as specified
1104 by other command line arguments.
1106 See tools/testing/selftests/mm/gup_test.c
1108 comment "GUP_TEST needs to have DEBUG_FS enabled"
1109 depends on !GUP_TEST && !DEBUG_FS
1111 config GUP_GET_PXX_LOW_HIGH
1115 tristate "Enable a module to run time tests on dma_pool"
1118 Provides a test module that will allocate and free many blocks of
1119 various sizes and report how long it takes. This is intended to
1120 provide a consistent way to measure how changes to the
1121 dma_pool_alloc/free routines affect performance.
1123 config ARCH_HAS_PTE_SPECIAL
1127 # Some architectures require a special hugepage directory format that is
1128 # required to support multiple hugepage sizes. For example a4fe3ce76
1129 # "powerpc/mm: Allow more flexible layouts for hugepage pagetables"
1130 # introduced it on powerpc. This allows for a more flexible hugepage
1131 # pagetable layouts.
1133 config ARCH_HAS_HUGEPD
1136 config MAPPING_DIRTY_HELPERS
1142 config KMAP_LOCAL_NON_LINEAR_PTE_ARRAY
1145 # struct io_mapping based helper. Selected by drivers that need them
1150 bool "Enable memfd_create() system call" if EXPERT
1154 bool "Enable memfd_secret() system call" if EXPERT
1155 depends on ARCH_HAS_SET_DIRECT_MAP
1157 Enable the memfd_secret() system call with the ability to create
1158 memory areas visible only in the context of the owning process and
1159 not mapped to other processes and other kernel page tables.
1161 config ANON_VMA_NAME
1162 bool "Anonymous VMA name support"
1163 depends on PROC_FS && ADVISE_SYSCALLS && MMU
1166 Allow naming anonymous virtual memory areas.
1168 This feature allows assigning names to virtual memory areas. Assigned
1169 names can be later retrieved from /proc/pid/maps and /proc/pid/smaps
1170 and help identifying individual anonymous memory areas.
1171 Assigning a name to anonymous virtual memory area might prevent that
1172 area from being merged with adjacent virtual memory areas due to the
1173 difference in their name.
1176 bool "Enable userfaultfd() system call"
1179 Enable the userfaultfd() system call that allows to intercept and
1180 handle page faults in userland.
1182 config HAVE_ARCH_USERFAULTFD_WP
1185 Arch has userfaultfd write protection support
1187 config HAVE_ARCH_USERFAULTFD_MINOR
1190 Arch has userfaultfd minor fault support
1192 config PTE_MARKER_UFFD_WP
1193 bool "Userfaultfd write protection support for shmem/hugetlbfs"
1195 depends on HAVE_ARCH_USERFAULTFD_WP
1198 Allows to create marker PTEs for userfaultfd write protection
1199 purposes. It is required to enable userfaultfd write protection on
1200 file-backed memory types like shmem and hugetlbfs.
1204 bool "Multi-Gen LRU"
1206 # make sure folio->flags has enough spare bits
1207 depends on 64BIT || !SPARSEMEM || SPARSEMEM_VMEMMAP
1209 A high performance LRU implementation to overcommit memory. See
1210 Documentation/admin-guide/mm/multigen_lru.rst for details.
1212 config LRU_GEN_ENABLED
1213 bool "Enable by default"
1216 This option enables the multi-gen LRU by default.
1218 config LRU_GEN_STATS
1219 bool "Full stats for debugging"
1222 Do not enable this option unless you plan to look at historical stats
1223 from evicted generations for debugging purpose.
1225 This option has a per-memcg and per-node memory overhead.
1228 config ARCH_SUPPORTS_PER_VMA_LOCK
1233 depends on ARCH_SUPPORTS_PER_VMA_LOCK && MMU && SMP
1235 Allow per-vma locking during page fault handling.
1237 This feature allows locking each virtual memory area separately when
1238 handling page faults instead of taking mmap_lock.
1240 config LOCK_MM_AND_FIND_VMA
1242 depends on !STACK_GROWSUP
1244 source "mm/damon/Kconfig"