1 config SELECT_MEMORY_MODEL
3 depends on EXPERIMENTAL || ARCH_SELECT_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
14 depends on !(ARCH_DISCONTIGMEM_ENABLE || ARCH_SPARSEMEM_ENABLE) || ARCH_FLATMEM_ENABLE
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
21 Some users of more advanced features like NUMA and
22 memory hotplug may have different options here.
23 DISCONTIGMEM is an 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".
29 If unsure, choose this option (Flat Memory) over any other.
31 config DISCONTIGMEM_MANUAL
32 bool "Discontiguous Memory"
33 depends on ARCH_DISCONTIGMEM_ENABLE
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
43 Many NUMA configurations will have this as the only option.
45 If unsure, choose "Flat Memory" over this option.
47 config SPARSEMEM_MANUAL
49 depends on ARCH_SPARSEMEM_ENABLE
51 This will be the only option for some systems, including
52 memory hotplug systems. This is normal.
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.
59 If unsure, choose "Discontiguous Memory" or "Flat Memory"
66 depends on (!SELECT_MEMORY_MODEL && ARCH_DISCONTIGMEM_ENABLE) || DISCONTIGMEM_MANUAL
70 depends on (!SELECT_MEMORY_MODEL && ARCH_SPARSEMEM_ENABLE) || SPARSEMEM_MANUAL
74 depends on (!DISCONTIGMEM && !SPARSEMEM) || FLATMEM_MANUAL
76 config FLAT_NODE_MEM_MAP
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.
85 config NEED_MULTIPLE_NODES
87 depends on DISCONTIGMEM || NUMA
89 config HAVE_MEMORY_PRESENT
91 depends on ARCH_HAVE_MEMORY_PRESENT || SPARSEMEM
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.
100 # This option will also potentially produce smaller runtime code
101 # with gcc 3.4 and later.
103 config SPARSEMEM_STATIC
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.
111 config SPARSEMEM_EXTREME
113 depends on SPARSEMEM && !SPARSEMEM_STATIC
115 config SPARSEMEM_VMEMMAP_ENABLE
118 config SPARSEMEM_ALLOC_MEM_MAP_TOGETHER
120 depends on SPARSEMEM && X86_64
122 config SPARSEMEM_VMEMMAP
123 bool "Sparse Memory virtual memmap"
124 depends on SPARSEMEM && SPARSEMEM_VMEMMAP_ENABLE
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.
134 config HAVE_MEMBLOCK_NODE_MAP
137 config ARCH_DISCARD_MEMBLOCK
143 config MEMORY_ISOLATION
147 boolean "Enable to assign a node which has only movable memory"
148 depends on HAVE_MEMBLOCK
149 depends on NO_BOOTMEM
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 users to
157 online all the memory of a node as movable memory so that the whole
158 node can be hotplugged. Users who don't use the memory hotplug
159 feature are fine with this option on since they don't online memory
162 Say Y here if you want to hotplug a whole node.
163 Say N here if you want kernel to use memory on all nodes evenly.
165 # eventually, we can have this option just 'select SPARSEMEM'
166 config MEMORY_HOTPLUG
167 bool "Allow for memory hot-add"
168 select MEMORY_ISOLATION
169 depends on SPARSEMEM || X86_64_ACPI_NUMA
170 depends on HOTPLUG && ARCH_ENABLE_MEMORY_HOTPLUG
171 depends on (IA64 || X86 || PPC_BOOK3S_64 || SUPERH || S390)
173 config MEMORY_HOTPLUG_SPARSE
175 depends on SPARSEMEM && MEMORY_HOTPLUG
177 config MEMORY_HOTREMOVE
178 bool "Allow for memory hot remove"
179 depends on MEMORY_HOTPLUG && ARCH_ENABLE_MEMORY_HOTREMOVE
183 # If we have space for more page flags then we can enable additional
184 # optimizations and functionality.
186 # Regular Sparsemem takes page flag bits for the sectionid if it does not
187 # use a virtual memmap. Disable extended page flags for 32 bit platforms
188 # that require the use of a sectionid in the page flags.
190 config PAGEFLAGS_EXTENDED
192 depends on 64BIT || SPARSEMEM_VMEMMAP || !SPARSEMEM
194 # Heavily threaded applications may benefit from splitting the mm-wide
195 # page_table_lock, so that faults on different parts of the user address
196 # space can be handled with less contention: split it at this NR_CPUS.
197 # Default to 4 for wider testing, though 8 might be more appropriate.
198 # ARM's adjust_pte (unused if VIPT) depends on mm-wide page_table_lock.
199 # PA-RISC 7xxx's spinlock_t would enlarge struct page from 32 to 44 bytes.
200 # DEBUG_SPINLOCK and DEBUG_LOCK_ALLOC spinlock_t also enlarge struct page.
202 config SPLIT_PTLOCK_CPUS
204 default "999999" if ARM && !CPU_CACHE_VIPT
205 default "999999" if PARISC && !PA20
206 default "999999" if DEBUG_SPINLOCK || DEBUG_LOCK_ALLOC
210 # support for memory balloon compaction
211 config BALLOON_COMPACTION
212 bool "Allow for balloon memory compaction/migration"
214 depends on COMPACTION && VIRTIO_BALLOON
216 Memory fragmentation introduced by ballooning might reduce
217 significantly the number of 2MB contiguous memory blocks that can be
218 used within a guest, thus imposing performance penalties associated
219 with the reduced number of transparent huge pages that could be used
220 by the guest workload. Allowing the compaction & migration for memory
221 pages enlisted as being part of memory balloon devices avoids the
222 scenario aforementioned and helps improving memory defragmentation.
225 # support for memory compaction
227 bool "Allow for memory compaction"
232 Allows the compaction of memory for the allocation of huge pages.
235 # support for page migration
238 bool "Page migration"
240 depends on NUMA || ARCH_ENABLE_MEMORY_HOTREMOVE || COMPACTION || CMA
242 Allows the migration of the physical location of pages of processes
243 while the virtual addresses are not changed. This is useful in
244 two situations. The first is on NUMA systems to put pages nearer
245 to the processors accessing. The second is when allocating huge
246 pages as migration can relocate pages to satisfy a huge page
247 allocation instead of reclaiming.
249 config PHYS_ADDR_T_64BIT
250 def_bool 64BIT || ARCH_PHYS_ADDR_T_64BIT
254 default "0" if !ZONE_DMA
259 depends on BLOCK && MMU && (ZONE_DMA || HIGHMEM)
269 depends on !ARCH_NO_VIRT_TO_BUS
275 bool "Enable KSM for page merging"
278 Enable Kernel Samepage Merging: KSM periodically scans those areas
279 of an application's address space that an app has advised may be
280 mergeable. When it finds pages of identical content, it replaces
281 the many instances by a single page with that content, so
282 saving memory until one or another app needs to modify the content.
283 Recommended for use with KVM, or with other duplicative applications.
284 See Documentation/vm/ksm.txt for more information: KSM is inactive
285 until a program has madvised that an area is MADV_MERGEABLE, and
286 root has set /sys/kernel/mm/ksm/run to 1 (if CONFIG_SYSFS is set).
288 config DEFAULT_MMAP_MIN_ADDR
289 int "Low address space to protect from user allocation"
293 This is the portion of low virtual memory which should be protected
294 from userspace allocation. Keeping a user from writing to low pages
295 can help reduce the impact of kernel NULL pointer bugs.
297 For most ia64, ppc64 and x86 users with lots of address space
298 a value of 65536 is reasonable and should cause no problems.
299 On arm and other archs it should not be higher than 32768.
300 Programs which use vm86 functionality or have some need to map
301 this low address space will need CAP_SYS_RAWIO or disable this
302 protection by setting the value to 0.
304 This value can be changed after boot using the
305 /proc/sys/vm/mmap_min_addr tunable.
307 config ARCH_SUPPORTS_MEMORY_FAILURE
310 config MEMORY_FAILURE
312 depends on ARCH_SUPPORTS_MEMORY_FAILURE
313 bool "Enable recovery from hardware memory errors"
314 select MEMORY_ISOLATION
316 Enables code to recover from some memory failures on systems
317 with MCA recovery. This allows a system to continue running
318 even when some of its memory has uncorrected errors. This requires
319 special hardware support and typically ECC memory.
321 config HWPOISON_INJECT
322 tristate "HWPoison pages injector"
323 depends on MEMORY_FAILURE && DEBUG_KERNEL && PROC_FS
324 select PROC_PAGE_MONITOR
326 config NOMMU_INITIAL_TRIM_EXCESS
327 int "Turn on mmap() excess space trimming before booting"
331 The NOMMU mmap() frequently needs to allocate large contiguous chunks
332 of memory on which to store mappings, but it can only ask the system
333 allocator for chunks in 2^N*PAGE_SIZE amounts - which is frequently
334 more than it requires. To deal with this, mmap() is able to trim off
335 the excess and return it to the allocator.
337 If trimming is enabled, the excess is trimmed off and returned to the
338 system allocator, which can cause extra fragmentation, particularly
339 if there are a lot of transient processes.
341 If trimming is disabled, the excess is kept, but not used, which for
342 long-term mappings means that the space is wasted.
344 Trimming can be dynamically controlled through a sysctl option
345 (/proc/sys/vm/nr_trim_pages) which specifies the minimum number of
346 excess pages there must be before trimming should occur, or zero if
347 no trimming is to occur.
349 This option specifies the initial value of this option. The default
350 of 1 says that all excess pages should be trimmed.
352 See Documentation/nommu-mmap.txt for more information.
354 config TRANSPARENT_HUGEPAGE
355 bool "Transparent Hugepage Support"
356 depends on HAVE_ARCH_TRANSPARENT_HUGEPAGE
359 Transparent Hugepages allows the kernel to use huge pages and
360 huge tlb transparently to the applications whenever possible.
361 This feature can improve computing performance to certain
362 applications by speeding up page faults during memory
363 allocation, by reducing the number of tlb misses and by speeding
364 up the pagetable walking.
366 If memory constrained on embedded, you may want to say N.
369 prompt "Transparent Hugepage Support sysfs defaults"
370 depends on TRANSPARENT_HUGEPAGE
371 default TRANSPARENT_HUGEPAGE_ALWAYS
373 Selects the sysfs defaults for Transparent Hugepage Support.
375 config TRANSPARENT_HUGEPAGE_ALWAYS
378 Enabling Transparent Hugepage always, can increase the
379 memory footprint of applications without a guaranteed
380 benefit but it will work automatically for all applications.
382 config TRANSPARENT_HUGEPAGE_MADVISE
385 Enabling Transparent Hugepage madvise, will only provide a
386 performance improvement benefit to the applications using
387 madvise(MADV_HUGEPAGE) but it won't risk to increase the
388 memory footprint of applications without a guaranteed
392 config CROSS_MEMORY_ATTACH
393 bool "Cross Memory Support"
397 Enabling this option adds the system calls process_vm_readv and
398 process_vm_writev which allow a process with the correct privileges
399 to directly read from or write to to another process's address space.
400 See the man page for more details.
403 # UP and nommu archs use km based percpu allocator
405 config NEED_PER_CPU_KM
411 bool "Enable cleancache driver to cache clean pages if tmem is present"
414 Cleancache can be thought of as a page-granularity victim cache
415 for clean pages that the kernel's pageframe replacement algorithm
416 (PFRA) would like to keep around, but can't since there isn't enough
417 memory. So when the PFRA "evicts" a page, it first attempts to use
418 cleancache code to put the data contained in that page into
419 "transcendent memory", memory that is not directly accessible or
420 addressable by the kernel and is of unknown and possibly
421 time-varying size. And when a cleancache-enabled
422 filesystem wishes to access a page in a file on disk, it first
423 checks cleancache to see if it already contains it; if it does,
424 the page is copied into the kernel and a disk access is avoided.
425 When a transcendent memory driver is available (such as zcache or
426 Xen transcendent memory), a significant I/O reduction
427 may be achieved. When none is available, all cleancache calls
428 are reduced to a single pointer-compare-against-NULL resulting
429 in a negligible performance hit.
431 If unsure, say Y to enable cleancache
434 bool "Enable frontswap to cache swap pages if tmem is present"
438 Frontswap is so named because it can be thought of as the opposite
439 of a "backing" store for a swap device. The data is stored into
440 "transcendent memory", memory that is not directly accessible or
441 addressable by the kernel and is of unknown and possibly
442 time-varying size. When space in transcendent memory is available,
443 a significant swap I/O reduction may be achieved. When none is
444 available, all frontswap calls are reduced to a single pointer-
445 compare-against-NULL resulting in a negligible performance hit
446 and swap data is stored as normal on the matching swap device.
448 If unsure, say Y to enable frontswap.