1config SELECT_MEMORY_MODEL 2 def_bool y 3 depends on ARCH_SELECT_MEMORY_MODEL 4 5choice 6 prompt "Memory model" 7 depends on SELECT_MEMORY_MODEL 8 default DISCONTIGMEM_MANUAL if ARCH_DISCONTIGMEM_DEFAULT 9 default SPARSEMEM_MANUAL if ARCH_SPARSEMEM_DEFAULT 10 default FLATMEM_MANUAL 11 12config FLATMEM_MANUAL 13 bool "Flat Memory" 14 depends on !(ARCH_DISCONTIGMEM_ENABLE || ARCH_SPARSEMEM_ENABLE) || ARCH_FLATMEM_ENABLE 15 help 16 This option allows you to change some of the ways that 17 Linux manages its memory internally. Most users will 18 only have one option here: FLATMEM. This is normal 19 and a correct option. 20 21 Some users of more advanced features like NUMA and 22 memory hotplug may have different options here. 23 DISCONTIGMEM is a more mature, better tested system, 24 but is incompatible with memory hotplug and may suffer 25 decreased performance over SPARSEMEM. If unsure between 26 "Sparse Memory" and "Discontiguous Memory", choose 27 "Discontiguous Memory". 28 29 If unsure, choose this option (Flat Memory) over any other. 30 31config DISCONTIGMEM_MANUAL 32 bool "Discontiguous Memory" 33 depends on ARCH_DISCONTIGMEM_ENABLE 34 help 35 This option provides enhanced support for discontiguous 36 memory systems, over FLATMEM. These systems have holes 37 in their physical address spaces, and this option provides 38 more efficient handling of these holes. However, the vast 39 majority of hardware has quite flat address spaces, and 40 can have degraded performance from the extra overhead that 41 this option imposes. 42 43 Many NUMA configurations will have this as the only option. 44 45 If unsure, choose "Flat Memory" over this option. 46 47config SPARSEMEM_MANUAL 48 bool "Sparse Memory" 49 depends on ARCH_SPARSEMEM_ENABLE 50 help 51 This will be the only option for some systems, including 52 memory hotplug systems. This is normal. 53 54 For many other systems, this will be an alternative to 55 "Discontiguous Memory". This option provides some potential 56 performance benefits, along with decreased code complexity, 57 but it is newer, and more experimental. 58 59 If unsure, choose "Discontiguous Memory" or "Flat Memory" 60 over this option. 61 62endchoice 63 64config DISCONTIGMEM 65 def_bool y 66 depends on (!SELECT_MEMORY_MODEL && ARCH_DISCONTIGMEM_ENABLE) || DISCONTIGMEM_MANUAL 67 68config SPARSEMEM 69 def_bool y 70 depends on (!SELECT_MEMORY_MODEL && ARCH_SPARSEMEM_ENABLE) || SPARSEMEM_MANUAL 71 72config FLATMEM 73 def_bool y 74 depends on (!DISCONTIGMEM && !SPARSEMEM) || FLATMEM_MANUAL 75 76config FLAT_NODE_MEM_MAP 77 def_bool y 78 depends on !SPARSEMEM 79 80# 81# Both the NUMA code and DISCONTIGMEM use arrays of pg_data_t's 82# to represent different areas of memory. This variable allows 83# those dependencies to exist individually. 84# 85config NEED_MULTIPLE_NODES 86 def_bool y 87 depends on DISCONTIGMEM || NUMA 88 89config HAVE_MEMORY_PRESENT 90 def_bool y 91 depends on ARCH_HAVE_MEMORY_PRESENT || SPARSEMEM 92 93# 94# SPARSEMEM_EXTREME (which is the default) does some bootmem 95# allocations when memory_present() is called. If this cannot 96# be done on your architecture, select this option. However, 97# statically allocating the mem_section[] array can potentially 98# consume vast quantities of .bss, so be careful. 99# 100# This option will also potentially produce smaller runtime code 101# with gcc 3.4 and later. 102# 103config SPARSEMEM_STATIC 104 bool 105 106# 107# Architecture platforms which require a two level mem_section in SPARSEMEM 108# must select this option. This is usually for architecture platforms with 109# an extremely sparse physical address space. 110# 111config SPARSEMEM_EXTREME 112 def_bool y 113 depends on SPARSEMEM && !SPARSEMEM_STATIC 114 115config SPARSEMEM_VMEMMAP_ENABLE 116 bool 117 118config SPARSEMEM_ALLOC_MEM_MAP_TOGETHER 119 def_bool y 120 depends on SPARSEMEM && X86_64 121 122config SPARSEMEM_VMEMMAP 123 bool "Sparse Memory virtual memmap" 124 depends on SPARSEMEM && SPARSEMEM_VMEMMAP_ENABLE 125 default y 126 help 127 SPARSEMEM_VMEMMAP uses a virtually mapped memmap to optimise 128 pfn_to_page and page_to_pfn operations. This is the most 129 efficient option when sufficient kernel resources are available. 130 131config HAVE_MEMBLOCK 132 bool 133 134config HAVE_MEMBLOCK_NODE_MAP 135 bool 136 137config HAVE_MEMBLOCK_PHYS_MAP 138 bool 139 140config HAVE_GENERIC_GUP 141 bool 142 143config ARCH_DISCARD_MEMBLOCK 144 bool 145 146config NO_BOOTMEM 147 bool 148 149config MEMORY_ISOLATION 150 bool 151 152# 153# Only be set on architectures that have completely implemented memory hotplug 154# feature. If you are not sure, don't touch it. 155# 156config HAVE_BOOTMEM_INFO_NODE 157 def_bool n 158 159# eventually, we can have this option just 'select SPARSEMEM' 160config MEMORY_HOTPLUG 161 bool "Allow for memory hot-add" 162 depends on SPARSEMEM || X86_64_ACPI_NUMA 163 depends on ARCH_ENABLE_MEMORY_HOTPLUG 164 depends on COMPILE_TEST || !KASAN 165 166config MEMORY_HOTPLUG_SPARSE 167 def_bool y 168 depends on SPARSEMEM && MEMORY_HOTPLUG 169 170config MEMORY_HOTPLUG_DEFAULT_ONLINE 171 bool "Online the newly added memory blocks by default" 172 default n 173 depends on MEMORY_HOTPLUG 174 help 175 This option sets the default policy setting for memory hotplug 176 onlining policy (/sys/devices/system/memory/auto_online_blocks) which 177 determines what happens to newly added memory regions. Policy setting 178 can always be changed at runtime. 179 See Documentation/memory-hotplug.txt for more information. 180 181 Say Y here if you want all hot-plugged memory blocks to appear in 182 'online' state by default. 183 Say N here if you want the default policy to keep all hot-plugged 184 memory blocks in 'offline' state. 185 186config MEMORY_HOTREMOVE 187 bool "Allow for memory hot remove" 188 select MEMORY_ISOLATION 189 select HAVE_BOOTMEM_INFO_NODE if (X86_64 || PPC64) 190 depends on MEMORY_HOTPLUG && ARCH_ENABLE_MEMORY_HOTREMOVE 191 depends on MIGRATION 192 193# Heavily threaded applications may benefit from splitting the mm-wide 194# page_table_lock, so that faults on different parts of the user address 195# space can be handled with less contention: split it at this NR_CPUS. 196# Default to 4 for wider testing, though 8 might be more appropriate. 197# ARM's adjust_pte (unused if VIPT) depends on mm-wide page_table_lock. 198# PA-RISC 7xxx's spinlock_t would enlarge struct page from 32 to 44 bytes. 199# DEBUG_SPINLOCK and DEBUG_LOCK_ALLOC spinlock_t also enlarge struct page. 200# 201config SPLIT_PTLOCK_CPUS 202 int 203 default "999999" if !MMU 204 default "999999" if ARM && !CPU_CACHE_VIPT 205 default "999999" if PARISC && !PA20 206 default "4" 207 208config ARCH_ENABLE_SPLIT_PMD_PTLOCK 209 bool 210 211# 212# support for memory balloon 213config MEMORY_BALLOON 214 bool 215 216# 217# support for memory balloon compaction 218config BALLOON_COMPACTION 219 bool "Allow for balloon memory compaction/migration" 220 def_bool y 221 depends on COMPACTION && MEMORY_BALLOON 222 help 223 Memory fragmentation introduced by ballooning might reduce 224 significantly the number of 2MB contiguous memory blocks that can be 225 used within a guest, thus imposing performance penalties associated 226 with the reduced number of transparent huge pages that could be used 227 by the guest workload. Allowing the compaction & migration for memory 228 pages enlisted as being part of memory balloon devices avoids the 229 scenario aforementioned and helps improving memory defragmentation. 230 231# 232# support for memory compaction 233config COMPACTION 234 bool "Allow for memory compaction" 235 def_bool y 236 select MIGRATION 237 depends on MMU 238 help 239 Compaction is the only memory management component to form 240 high order (larger physically contiguous) memory blocks 241 reliably. The page allocator relies on compaction heavily and 242 the lack of the feature can lead to unexpected OOM killer 243 invocations for high order memory requests. You shouldn't 244 disable this option unless there really is a strong reason for 245 it and then we would be really interested to hear about that at 246 linux-mm@kvack.org. 247 248# 249# support for page migration 250# 251config MIGRATION 252 bool "Page migration" 253 def_bool y 254 depends on (NUMA || ARCH_ENABLE_MEMORY_HOTREMOVE || COMPACTION || CMA) && MMU 255 help 256 Allows the migration of the physical location of pages of processes 257 while the virtual addresses are not changed. This is useful in 258 two situations. The first is on NUMA systems to put pages nearer 259 to the processors accessing. The second is when allocating huge 260 pages as migration can relocate pages to satisfy a huge page 261 allocation instead of reclaiming. 262 263config ARCH_ENABLE_HUGEPAGE_MIGRATION 264 bool 265 266config PHYS_ADDR_T_64BIT 267 def_bool 64BIT || ARCH_PHYS_ADDR_T_64BIT 268 269config BOUNCE 270 bool "Enable bounce buffers" 271 default y 272 depends on BLOCK && MMU && (ZONE_DMA || HIGHMEM) 273 help 274 Enable bounce buffers for devices that cannot access 275 the full range of memory available to the CPU. Enabled 276 by default when ZONE_DMA or HIGHMEM is selected, but you 277 may say n to override this. 278 279# On the 'tile' arch, USB OHCI needs the bounce pool since tilegx will often 280# have more than 4GB of memory, but we don't currently use the IOTLB to present 281# a 32-bit address to OHCI. So we need to use a bounce pool instead. 282config NEED_BOUNCE_POOL 283 bool 284 default y if TILE && USB_OHCI_HCD 285 286config NR_QUICK 287 int 288 depends on QUICKLIST 289 default "1" 290 291config VIRT_TO_BUS 292 bool 293 help 294 An architecture should select this if it implements the 295 deprecated interface virt_to_bus(). All new architectures 296 should probably not select this. 297 298 299config MMU_NOTIFIER 300 bool 301 select SRCU 302 303config KSM 304 bool "Enable KSM for page merging" 305 depends on MMU 306 help 307 Enable Kernel Samepage Merging: KSM periodically scans those areas 308 of an application's address space that an app has advised may be 309 mergeable. When it finds pages of identical content, it replaces 310 the many instances by a single page with that content, so 311 saving memory until one or another app needs to modify the content. 312 Recommended for use with KVM, or with other duplicative applications. 313 See Documentation/vm/ksm.txt for more information: KSM is inactive 314 until a program has madvised that an area is MADV_MERGEABLE, and 315 root has set /sys/kernel/mm/ksm/run to 1 (if CONFIG_SYSFS is set). 316 317config DEFAULT_MMAP_MIN_ADDR 318 int "Low address space to protect from user allocation" 319 depends on MMU 320 default 4096 321 help 322 This is the portion of low virtual memory which should be protected 323 from userspace allocation. Keeping a user from writing to low pages 324 can help reduce the impact of kernel NULL pointer bugs. 325 326 For most ia64, ppc64 and x86 users with lots of address space 327 a value of 65536 is reasonable and should cause no problems. 328 On arm and other archs it should not be higher than 32768. 329 Programs which use vm86 functionality or have some need to map 330 this low address space will need CAP_SYS_RAWIO or disable this 331 protection by setting the value to 0. 332 333 This value can be changed after boot using the 334 /proc/sys/vm/mmap_min_addr tunable. 335 336config ARCH_SUPPORTS_MEMORY_FAILURE 337 bool 338 339config MEMORY_FAILURE 340 depends on MMU 341 depends on ARCH_SUPPORTS_MEMORY_FAILURE 342 bool "Enable recovery from hardware memory errors" 343 select MEMORY_ISOLATION 344 select RAS 345 help 346 Enables code to recover from some memory failures on systems 347 with MCA recovery. This allows a system to continue running 348 even when some of its memory has uncorrected errors. This requires 349 special hardware support and typically ECC memory. 350 351config HWPOISON_INJECT 352 tristate "HWPoison pages injector" 353 depends on MEMORY_FAILURE && DEBUG_KERNEL && PROC_FS 354 select PROC_PAGE_MONITOR 355 356config NOMMU_INITIAL_TRIM_EXCESS 357 int "Turn on mmap() excess space trimming before booting" 358 depends on !MMU 359 default 1 360 help 361 The NOMMU mmap() frequently needs to allocate large contiguous chunks 362 of memory on which to store mappings, but it can only ask the system 363 allocator for chunks in 2^N*PAGE_SIZE amounts - which is frequently 364 more than it requires. To deal with this, mmap() is able to trim off 365 the excess and return it to the allocator. 366 367 If trimming is enabled, the excess is trimmed off and returned to the 368 system allocator, which can cause extra fragmentation, particularly 369 if there are a lot of transient processes. 370 371 If trimming is disabled, the excess is kept, but not used, which for 372 long-term mappings means that the space is wasted. 373 374 Trimming can be dynamically controlled through a sysctl option 375 (/proc/sys/vm/nr_trim_pages) which specifies the minimum number of 376 excess pages there must be before trimming should occur, or zero if 377 no trimming is to occur. 378 379 This option specifies the initial value of this option. The default 380 of 1 says that all excess pages should be trimmed. 381 382 See Documentation/nommu-mmap.txt for more information. 383 384config TRANSPARENT_HUGEPAGE 385 bool "Transparent Hugepage Support" 386 depends on HAVE_ARCH_TRANSPARENT_HUGEPAGE 387 select COMPACTION 388 select RADIX_TREE_MULTIORDER 389 help 390 Transparent Hugepages allows the kernel to use huge pages and 391 huge tlb transparently to the applications whenever possible. 392 This feature can improve computing performance to certain 393 applications by speeding up page faults during memory 394 allocation, by reducing the number of tlb misses and by speeding 395 up the pagetable walking. 396 397 If memory constrained on embedded, you may want to say N. 398 399choice 400 prompt "Transparent Hugepage Support sysfs defaults" 401 depends on TRANSPARENT_HUGEPAGE 402 default TRANSPARENT_HUGEPAGE_ALWAYS 403 help 404 Selects the sysfs defaults for Transparent Hugepage Support. 405 406 config TRANSPARENT_HUGEPAGE_ALWAYS 407 bool "always" 408 help 409 Enabling Transparent Hugepage always, can increase the 410 memory footprint of applications without a guaranteed 411 benefit but it will work automatically for all applications. 412 413 config TRANSPARENT_HUGEPAGE_MADVISE 414 bool "madvise" 415 help 416 Enabling Transparent Hugepage madvise, will only provide a 417 performance improvement benefit to the applications using 418 madvise(MADV_HUGEPAGE) but it won't risk to increase the 419 memory footprint of applications without a guaranteed 420 benefit. 421endchoice 422 423config ARCH_WANTS_THP_SWAP 424 def_bool n 425 426config THP_SWAP 427 def_bool y 428 depends on TRANSPARENT_HUGEPAGE && ARCH_WANTS_THP_SWAP 429 help 430 Swap transparent huge pages in one piece, without splitting. 431 XXX: For now this only does clustered swap space allocation. 432 433 For selection by architectures with reasonable THP sizes. 434 435config TRANSPARENT_HUGE_PAGECACHE 436 def_bool y 437 depends on TRANSPARENT_HUGEPAGE 438 439# 440# UP and nommu archs use km based percpu allocator 441# 442config NEED_PER_CPU_KM 443 depends on !SMP 444 bool 445 default y 446 447config CLEANCACHE 448 bool "Enable cleancache driver to cache clean pages if tmem is present" 449 default n 450 help 451 Cleancache can be thought of as a page-granularity victim cache 452 for clean pages that the kernel's pageframe replacement algorithm 453 (PFRA) would like to keep around, but can't since there isn't enough 454 memory. So when the PFRA "evicts" a page, it first attempts to use 455 cleancache code to put the data contained in that page into 456 "transcendent memory", memory that is not directly accessible or 457 addressable by the kernel and is of unknown and possibly 458 time-varying size. And when a cleancache-enabled 459 filesystem wishes to access a page in a file on disk, it first 460 checks cleancache to see if it already contains it; if it does, 461 the page is copied into the kernel and a disk access is avoided. 462 When a transcendent memory driver is available (such as zcache or 463 Xen transcendent memory), a significant I/O reduction 464 may be achieved. When none is available, all cleancache calls 465 are reduced to a single pointer-compare-against-NULL resulting 466 in a negligible performance hit. 467 468 If unsure, say Y to enable cleancache 469 470config FRONTSWAP 471 bool "Enable frontswap to cache swap pages if tmem is present" 472 depends on SWAP 473 default n 474 help 475 Frontswap is so named because it can be thought of as the opposite 476 of a "backing" store for a swap device. The data is stored into 477 "transcendent memory", memory that is not directly accessible or 478 addressable by the kernel and is of unknown and possibly 479 time-varying size. When space in transcendent memory is available, 480 a significant swap I/O reduction may be achieved. When none is 481 available, all frontswap calls are reduced to a single pointer- 482 compare-against-NULL resulting in a negligible performance hit 483 and swap data is stored as normal on the matching swap device. 484 485 If unsure, say Y to enable frontswap. 486 487config CMA 488 bool "Contiguous Memory Allocator" 489 depends on HAVE_MEMBLOCK && MMU 490 select MIGRATION 491 select MEMORY_ISOLATION 492 help 493 This enables the Contiguous Memory Allocator which allows other 494 subsystems to allocate big physically-contiguous blocks of memory. 495 CMA reserves a region of memory and allows only movable pages to 496 be allocated from it. This way, the kernel can use the memory for 497 pagecache and when a subsystem requests for contiguous area, the 498 allocated pages are migrated away to serve the contiguous request. 499 500 If unsure, say "n". 501 502config CMA_DEBUG 503 bool "CMA debug messages (DEVELOPMENT)" 504 depends on DEBUG_KERNEL && CMA 505 help 506 Turns on debug messages in CMA. This produces KERN_DEBUG 507 messages for every CMA call as well as various messages while 508 processing calls such as dma_alloc_from_contiguous(). 509 This option does not affect warning and error messages. 510 511config CMA_DEBUGFS 512 bool "CMA debugfs interface" 513 depends on CMA && DEBUG_FS 514 help 515 Turns on the DebugFS interface for CMA. 516 517config CMA_AREAS 518 int "Maximum count of the CMA areas" 519 depends on CMA 520 default 7 521 help 522 CMA allows to create CMA areas for particular purpose, mainly, 523 used as device private area. This parameter sets the maximum 524 number of CMA area in the system. 525 526 If unsure, leave the default value "7". 527 528config MEM_SOFT_DIRTY 529 bool "Track memory changes" 530 depends on CHECKPOINT_RESTORE && HAVE_ARCH_SOFT_DIRTY && PROC_FS 531 select PROC_PAGE_MONITOR 532 help 533 This option enables memory changes tracking by introducing a 534 soft-dirty bit on pte-s. This bit it set when someone writes 535 into a page just as regular dirty bit, but unlike the latter 536 it can be cleared by hands. 537 538 See Documentation/vm/soft-dirty.txt for more details. 539 540config ZSWAP 541 bool "Compressed cache for swap pages (EXPERIMENTAL)" 542 depends on FRONTSWAP && CRYPTO=y 543 select CRYPTO_LZO 544 select ZPOOL 545 default n 546 help 547 A lightweight compressed cache for swap pages. It takes 548 pages that are in the process of being swapped out and attempts to 549 compress them into a dynamically allocated RAM-based memory pool. 550 This can result in a significant I/O reduction on swap device and, 551 in the case where decompressing from RAM is faster that swap device 552 reads, can also improve workload performance. 553 554 This is marked experimental because it is a new feature (as of 555 v3.11) that interacts heavily with memory reclaim. While these 556 interactions don't cause any known issues on simple memory setups, 557 they have not be fully explored on the large set of potential 558 configurations and workloads that exist. 559 560config ZPOOL 561 tristate "Common API for compressed memory storage" 562 default n 563 help 564 Compressed memory storage API. This allows using either zbud or 565 zsmalloc. 566 567config ZBUD 568 tristate "Low (Up to 2x) density storage for compressed pages" 569 default n 570 help 571 A special purpose allocator for storing compressed pages. 572 It is designed to store up to two compressed pages per physical 573 page. While this design limits storage density, it has simple and 574 deterministic reclaim properties that make it preferable to a higher 575 density approach when reclaim will be used. 576 577config Z3FOLD 578 tristate "Up to 3x density storage for compressed pages" 579 depends on ZPOOL 580 default n 581 help 582 A special purpose allocator for storing compressed pages. 583 It is designed to store up to three compressed pages per physical 584 page. It is a ZBUD derivative so the simplicity and determinism are 585 still there. 586 587config ZSMALLOC 588 tristate "Memory allocator for compressed pages" 589 depends on MMU 590 default n 591 help 592 zsmalloc is a slab-based memory allocator designed to store 593 compressed RAM pages. zsmalloc uses virtual memory mapping 594 in order to reduce fragmentation. However, this results in a 595 non-standard allocator interface where a handle, not a pointer, is 596 returned by an alloc(). This handle must be mapped in order to 597 access the allocated space. 598 599config PGTABLE_MAPPING 600 bool "Use page table mapping to access object in zsmalloc" 601 depends on ZSMALLOC 602 help 603 By default, zsmalloc uses a copy-based object mapping method to 604 access allocations that span two pages. However, if a particular 605 architecture (ex, ARM) performs VM mapping faster than copying, 606 then you should select this. This causes zsmalloc to use page table 607 mapping rather than copying for object mapping. 608 609 You can check speed with zsmalloc benchmark: 610 https://github.com/spartacus06/zsmapbench 611 612config ZSMALLOC_STAT 613 bool "Export zsmalloc statistics" 614 depends on ZSMALLOC 615 select DEBUG_FS 616 help 617 This option enables code in the zsmalloc to collect various 618 statistics about whats happening in zsmalloc and exports that 619 information to userspace via debugfs. 620 If unsure, say N. 621 622config GENERIC_EARLY_IOREMAP 623 bool 624 625config MAX_STACK_SIZE_MB 626 int "Maximum user stack size for 32-bit processes (MB)" 627 default 80 628 range 8 256 if METAG 629 range 8 2048 630 depends on STACK_GROWSUP && (!64BIT || COMPAT) 631 help 632 This is the maximum stack size in Megabytes in the VM layout of 32-bit 633 user processes when the stack grows upwards (currently only on parisc 634 and metag arch). The stack will be located at the highest memory 635 address minus the given value, unless the RLIMIT_STACK hard limit is 636 changed to a smaller value in which case that is used. 637 638 A sane initial value is 80 MB. 639 640# For architectures that support deferred memory initialisation 641config ARCH_SUPPORTS_DEFERRED_STRUCT_PAGE_INIT 642 bool 643 644config DEFERRED_STRUCT_PAGE_INIT 645 bool "Defer initialisation of struct pages to kthreads" 646 default n 647 depends on ARCH_SUPPORTS_DEFERRED_STRUCT_PAGE_INIT 648 depends on NO_BOOTMEM && MEMORY_HOTPLUG 649 depends on !FLATMEM 650 help 651 Ordinarily all struct pages are initialised during early boot in a 652 single thread. On very large machines this can take a considerable 653 amount of time. If this option is set, large machines will bring up 654 a subset of memmap at boot and then initialise the rest in parallel 655 by starting one-off "pgdatinitX" kernel thread for each node X. This 656 has a potential performance impact on processes running early in the 657 lifetime of the system until these kthreads finish the 658 initialisation. 659 660config IDLE_PAGE_TRACKING 661 bool "Enable idle page tracking" 662 depends on SYSFS && MMU 663 select PAGE_EXTENSION if !64BIT 664 help 665 This feature allows to estimate the amount of user pages that have 666 not been touched during a given period of time. This information can 667 be useful to tune memory cgroup limits and/or for job placement 668 within a compute cluster. 669 670 See Documentation/vm/idle_page_tracking.txt for more details. 671 672# arch_add_memory() comprehends device memory 673config ARCH_HAS_ZONE_DEVICE 674 bool 675 676config ZONE_DEVICE 677 bool "Device memory (pmem, etc...) hotplug support" 678 depends on MEMORY_HOTPLUG 679 depends on MEMORY_HOTREMOVE 680 depends on SPARSEMEM_VMEMMAP 681 depends on ARCH_HAS_ZONE_DEVICE 682 683 help 684 Device memory hotplug support allows for establishing pmem, 685 or other device driver discovered memory regions, in the 686 memmap. This allows pfn_to_page() lookups of otherwise 687 "device-physical" addresses which is needed for using a DAX 688 mapping in an O_DIRECT operation, among other things. 689 690 If FS_DAX is enabled, then say Y. 691 692config FRAME_VECTOR 693 bool 694 695config ARCH_USES_HIGH_VMA_FLAGS 696 bool 697config ARCH_HAS_PKEYS 698 bool 699 700config PERCPU_STATS 701 bool "Collect percpu memory statistics" 702 default n 703 help 704 This feature collects and exposes statistics via debugfs. The 705 information includes global and per chunk statistics, which can 706 be used to help understand percpu memory usage. 707