1e1785e85SDave Hansenconfig SELECT_MEMORY_MODEL 2e1785e85SDave Hansen def_bool y 3a8826eebSKees Cook depends on ARCH_SELECT_MEMORY_MODEL 4e1785e85SDave Hansen 53a9da765SDave Hansenchoice 63a9da765SDave Hansen prompt "Memory model" 7e1785e85SDave Hansen depends on SELECT_MEMORY_MODEL 8e1785e85SDave Hansen default DISCONTIGMEM_MANUAL if ARCH_DISCONTIGMEM_DEFAULT 9d41dee36SAndy Whitcroft default SPARSEMEM_MANUAL if ARCH_SPARSEMEM_DEFAULT 10e1785e85SDave Hansen default FLATMEM_MANUAL 113a9da765SDave Hansen 12e1785e85SDave Hansenconfig FLATMEM_MANUAL 133a9da765SDave Hansen bool "Flat Memory" 14c898ec16SAnton Blanchard depends on !(ARCH_DISCONTIGMEM_ENABLE || ARCH_SPARSEMEM_ENABLE) || ARCH_FLATMEM_ENABLE 153a9da765SDave Hansen help 163a9da765SDave Hansen This option allows you to change some of the ways that 173a9da765SDave Hansen Linux manages its memory internally. Most users will 183a9da765SDave Hansen only have one option here: FLATMEM. This is normal 193a9da765SDave Hansen and a correct option. 203a9da765SDave Hansen 21d41dee36SAndy Whitcroft Some users of more advanced features like NUMA and 22d41dee36SAndy Whitcroft memory hotplug may have different options here. 2318f65332SGeert Uytterhoeven DISCONTIGMEM is a more mature, better tested system, 24d41dee36SAndy Whitcroft but is incompatible with memory hotplug and may suffer 25d41dee36SAndy Whitcroft decreased performance over SPARSEMEM. If unsure between 26d41dee36SAndy Whitcroft "Sparse Memory" and "Discontiguous Memory", choose 27d41dee36SAndy Whitcroft "Discontiguous Memory". 28d41dee36SAndy Whitcroft 29d41dee36SAndy Whitcroft If unsure, choose this option (Flat Memory) over any other. 303a9da765SDave Hansen 31e1785e85SDave Hansenconfig DISCONTIGMEM_MANUAL 32f3519f91SDave Hansen bool "Discontiguous Memory" 333a9da765SDave Hansen depends on ARCH_DISCONTIGMEM_ENABLE 343a9da765SDave Hansen help 35785dcd44SDave Hansen This option provides enhanced support for discontiguous 36785dcd44SDave Hansen memory systems, over FLATMEM. These systems have holes 37785dcd44SDave Hansen in their physical address spaces, and this option provides 38785dcd44SDave Hansen more efficient handling of these holes. However, the vast 39785dcd44SDave Hansen majority of hardware has quite flat address spaces, and 40ad3d0a38SPhilipp Marek can have degraded performance from the extra overhead that 41785dcd44SDave Hansen this option imposes. 42785dcd44SDave Hansen 43785dcd44SDave Hansen Many NUMA configurations will have this as the only option. 44785dcd44SDave Hansen 453a9da765SDave Hansen If unsure, choose "Flat Memory" over this option. 463a9da765SDave Hansen 47d41dee36SAndy Whitcroftconfig SPARSEMEM_MANUAL 48d41dee36SAndy Whitcroft bool "Sparse Memory" 49d41dee36SAndy Whitcroft depends on ARCH_SPARSEMEM_ENABLE 50d41dee36SAndy Whitcroft help 51d41dee36SAndy Whitcroft This will be the only option for some systems, including 52d41dee36SAndy Whitcroft memory hotplug systems. This is normal. 53d41dee36SAndy Whitcroft 54d41dee36SAndy Whitcroft For many other systems, this will be an alternative to 55f3519f91SDave Hansen "Discontiguous Memory". This option provides some potential 56d41dee36SAndy Whitcroft performance benefits, along with decreased code complexity, 57d41dee36SAndy Whitcroft but it is newer, and more experimental. 58d41dee36SAndy Whitcroft 59d41dee36SAndy Whitcroft If unsure, choose "Discontiguous Memory" or "Flat Memory" 60d41dee36SAndy Whitcroft over this option. 61d41dee36SAndy Whitcroft 623a9da765SDave Hansenendchoice 633a9da765SDave Hansen 64e1785e85SDave Hansenconfig DISCONTIGMEM 65e1785e85SDave Hansen def_bool y 66e1785e85SDave Hansen depends on (!SELECT_MEMORY_MODEL && ARCH_DISCONTIGMEM_ENABLE) || DISCONTIGMEM_MANUAL 67e1785e85SDave Hansen 68d41dee36SAndy Whitcroftconfig SPARSEMEM 69d41dee36SAndy Whitcroft def_bool y 701a83e175SRussell King depends on (!SELECT_MEMORY_MODEL && ARCH_SPARSEMEM_ENABLE) || SPARSEMEM_MANUAL 71d41dee36SAndy Whitcroft 72e1785e85SDave Hansenconfig FLATMEM 73e1785e85SDave Hansen def_bool y 74d41dee36SAndy Whitcroft depends on (!DISCONTIGMEM && !SPARSEMEM) || FLATMEM_MANUAL 75d41dee36SAndy Whitcroft 76d41dee36SAndy Whitcroftconfig FLAT_NODE_MEM_MAP 77d41dee36SAndy Whitcroft def_bool y 78d41dee36SAndy Whitcroft depends on !SPARSEMEM 79e1785e85SDave Hansen 8093b7504eSDave Hansen# 8193b7504eSDave Hansen# Both the NUMA code and DISCONTIGMEM use arrays of pg_data_t's 8293b7504eSDave Hansen# to represent different areas of memory. This variable allows 8393b7504eSDave Hansen# those dependencies to exist individually. 8493b7504eSDave Hansen# 8593b7504eSDave Hansenconfig NEED_MULTIPLE_NODES 8693b7504eSDave Hansen def_bool y 8793b7504eSDave Hansen depends on DISCONTIGMEM || NUMA 88af705362SAndy Whitcroft 89af705362SAndy Whitcroftconfig HAVE_MEMORY_PRESENT 90af705362SAndy Whitcroft def_bool y 91d41dee36SAndy Whitcroft depends on ARCH_HAVE_MEMORY_PRESENT || SPARSEMEM 92802f192eSBob Picco 93802f192eSBob Picco# 943e347261SBob Picco# SPARSEMEM_EXTREME (which is the default) does some bootmem 953e347261SBob Picco# allocations when memory_present() is called. If this cannot 963e347261SBob Picco# be done on your architecture, select this option. However, 973e347261SBob Picco# statically allocating the mem_section[] array can potentially 983e347261SBob Picco# consume vast quantities of .bss, so be careful. 993e347261SBob Picco# 1003e347261SBob Picco# This option will also potentially produce smaller runtime code 1013e347261SBob Picco# with gcc 3.4 and later. 1023e347261SBob Picco# 1033e347261SBob Piccoconfig SPARSEMEM_STATIC 1049ba16087SJan Beulich bool 1053e347261SBob Picco 1063e347261SBob Picco# 10744c09201SMatt LaPlante# Architecture platforms which require a two level mem_section in SPARSEMEM 108802f192eSBob Picco# must select this option. This is usually for architecture platforms with 109802f192eSBob Picco# an extremely sparse physical address space. 110802f192eSBob Picco# 1113e347261SBob Piccoconfig SPARSEMEM_EXTREME 1123e347261SBob Picco def_bool y 1133e347261SBob Picco depends on SPARSEMEM && !SPARSEMEM_STATIC 1144c21e2f2SHugh Dickins 11529c71111SAndy Whitcroftconfig SPARSEMEM_VMEMMAP_ENABLE 1169ba16087SJan Beulich bool 11729c71111SAndy Whitcroft 1189bdac914SYinghai Luconfig SPARSEMEM_ALLOC_MEM_MAP_TOGETHER 1199bdac914SYinghai Lu def_bool y 1209bdac914SYinghai Lu depends on SPARSEMEM && X86_64 1219bdac914SYinghai Lu 12229c71111SAndy Whitcroftconfig SPARSEMEM_VMEMMAP 123a5ee6daaSGeoff Levand bool "Sparse Memory virtual memmap" 124a5ee6daaSGeoff Levand depends on SPARSEMEM && SPARSEMEM_VMEMMAP_ENABLE 125a5ee6daaSGeoff Levand default y 126a5ee6daaSGeoff Levand help 127a5ee6daaSGeoff Levand SPARSEMEM_VMEMMAP uses a virtually mapped memmap to optimise 128a5ee6daaSGeoff Levand pfn_to_page and page_to_pfn operations. This is the most 129a5ee6daaSGeoff Levand efficient option when sufficient kernel resources are available. 13029c71111SAndy Whitcroft 13195f72d1eSYinghai Luconfig HAVE_MEMBLOCK 13295f72d1eSYinghai Lu boolean 13395f72d1eSYinghai Lu 1347c0caeb8STejun Heoconfig HAVE_MEMBLOCK_NODE_MAP 1357c0caeb8STejun Heo boolean 1367c0caeb8STejun Heo 137c378ddd5STejun Heoconfig ARCH_DISCARD_MEMBLOCK 138c378ddd5STejun Heo boolean 139c378ddd5STejun Heo 14066616720SSam Ravnborgconfig NO_BOOTMEM 14166616720SSam Ravnborg boolean 14266616720SSam Ravnborg 143ee6f509cSMinchan Kimconfig MEMORY_ISOLATION 144ee6f509cSMinchan Kim boolean 145ee6f509cSMinchan Kim 14620b2f52bSLai Jiangshanconfig MOVABLE_NODE 14720b2f52bSLai Jiangshan boolean "Enable to assign a node which has only movable memory" 14820b2f52bSLai Jiangshan depends on HAVE_MEMBLOCK 14920b2f52bSLai Jiangshan depends on NO_BOOTMEM 15020b2f52bSLai Jiangshan depends on X86_64 15120b2f52bSLai Jiangshan depends on NUMA 152c2974058STang Chen default n 153c2974058STang Chen help 154c2974058STang Chen Allow a node to have only movable memory. Pages used by the kernel, 155c2974058STang Chen such as direct mapping pages cannot be migrated. So the corresponding 156c2974058STang Chen memory device cannot be hotplugged. This option allows users to 157c2974058STang Chen online all the memory of a node as movable memory so that the whole 158c2974058STang Chen node can be hotplugged. Users who don't use the memory hotplug 159c2974058STang Chen feature are fine with this option on since they don't online memory 160c2974058STang Chen as movable. 161c2974058STang Chen 162c2974058STang Chen Say Y here if you want to hotplug a whole node. 163c2974058STang Chen Say N here if you want kernel to use memory on all nodes evenly. 16420b2f52bSLai Jiangshan 16546723bfaSYasuaki Ishimatsu# 16646723bfaSYasuaki Ishimatsu# Only be set on architectures that have completely implemented memory hotplug 16746723bfaSYasuaki Ishimatsu# feature. If you are not sure, don't touch it. 16846723bfaSYasuaki Ishimatsu# 16946723bfaSYasuaki Ishimatsuconfig HAVE_BOOTMEM_INFO_NODE 17046723bfaSYasuaki Ishimatsu def_bool n 17146723bfaSYasuaki Ishimatsu 1723947be19SDave Hansen# eventually, we can have this option just 'select SPARSEMEM' 1733947be19SDave Hansenconfig MEMORY_HOTPLUG 1743947be19SDave Hansen bool "Allow for memory hot-add" 175ec69acbbSKeith Mannthey depends on SPARSEMEM || X86_64_ACPI_NUMA 17640b31360SStephen Rothwell depends on ARCH_ENABLE_MEMORY_HOTPLUG 177ed84a07aSKumar Gala depends on (IA64 || X86 || PPC_BOOK3S_64 || SUPERH || S390) 1783947be19SDave Hansen 179ec69acbbSKeith Manntheyconfig MEMORY_HOTPLUG_SPARSE 180ec69acbbSKeith Mannthey def_bool y 181ec69acbbSKeith Mannthey depends on SPARSEMEM && MEMORY_HOTPLUG 182ec69acbbSKeith Mannthey 1830c0e6195SKAMEZAWA Hiroyukiconfig MEMORY_HOTREMOVE 1840c0e6195SKAMEZAWA Hiroyuki bool "Allow for memory hot remove" 18546723bfaSYasuaki Ishimatsu select MEMORY_ISOLATION 18646723bfaSYasuaki Ishimatsu select HAVE_BOOTMEM_INFO_NODE if X86_64 1870c0e6195SKAMEZAWA Hiroyuki depends on MEMORY_HOTPLUG && ARCH_ENABLE_MEMORY_HOTREMOVE 1880c0e6195SKAMEZAWA Hiroyuki depends on MIGRATION 1890c0e6195SKAMEZAWA Hiroyuki 190e20b8ccaSChristoph Lameter# 191e20b8ccaSChristoph Lameter# If we have space for more page flags then we can enable additional 192e20b8ccaSChristoph Lameter# optimizations and functionality. 193e20b8ccaSChristoph Lameter# 194e20b8ccaSChristoph Lameter# Regular Sparsemem takes page flag bits for the sectionid if it does not 195e20b8ccaSChristoph Lameter# use a virtual memmap. Disable extended page flags for 32 bit platforms 196e20b8ccaSChristoph Lameter# that require the use of a sectionid in the page flags. 197e20b8ccaSChristoph Lameter# 198e20b8ccaSChristoph Lameterconfig PAGEFLAGS_EXTENDED 199e20b8ccaSChristoph Lameter def_bool y 200a269cca9SH. Peter Anvin depends on 64BIT || SPARSEMEM_VMEMMAP || !SPARSEMEM 201e20b8ccaSChristoph Lameter 2024c21e2f2SHugh Dickins# Heavily threaded applications may benefit from splitting the mm-wide 2034c21e2f2SHugh Dickins# page_table_lock, so that faults on different parts of the user address 2044c21e2f2SHugh Dickins# space can be handled with less contention: split it at this NR_CPUS. 2054c21e2f2SHugh Dickins# Default to 4 for wider testing, though 8 might be more appropriate. 2064c21e2f2SHugh Dickins# ARM's adjust_pte (unused if VIPT) depends on mm-wide page_table_lock. 2077b6ac9dfSHugh Dickins# PA-RISC 7xxx's spinlock_t would enlarge struct page from 32 to 44 bytes. 208a70caa8bSHugh Dickins# DEBUG_SPINLOCK and DEBUG_LOCK_ALLOC spinlock_t also enlarge struct page. 2094c21e2f2SHugh Dickins# 2104c21e2f2SHugh Dickinsconfig SPLIT_PTLOCK_CPUS 2114c21e2f2SHugh Dickins int 212a70caa8bSHugh Dickins default "999999" if ARM && !CPU_CACHE_VIPT 213a70caa8bSHugh Dickins default "999999" if PARISC && !PA20 214a70caa8bSHugh Dickins default "999999" if DEBUG_SPINLOCK || DEBUG_LOCK_ALLOC 2154c21e2f2SHugh Dickins default "4" 2167cbe34cfSChristoph Lameter 2177cbe34cfSChristoph Lameter# 21818468d93SRafael Aquini# support for memory balloon compaction 21918468d93SRafael Aquiniconfig BALLOON_COMPACTION 22018468d93SRafael Aquini bool "Allow for balloon memory compaction/migration" 22118468d93SRafael Aquini def_bool y 22218468d93SRafael Aquini depends on COMPACTION && VIRTIO_BALLOON 22318468d93SRafael Aquini help 22418468d93SRafael Aquini Memory fragmentation introduced by ballooning might reduce 22518468d93SRafael Aquini significantly the number of 2MB contiguous memory blocks that can be 22618468d93SRafael Aquini used within a guest, thus imposing performance penalties associated 22718468d93SRafael Aquini with the reduced number of transparent huge pages that could be used 22818468d93SRafael Aquini by the guest workload. Allowing the compaction & migration for memory 22918468d93SRafael Aquini pages enlisted as being part of memory balloon devices avoids the 23018468d93SRafael Aquini scenario aforementioned and helps improving memory defragmentation. 23118468d93SRafael Aquini 23218468d93SRafael Aquini# 233e9e96b39SMel Gorman# support for memory compaction 234e9e96b39SMel Gormanconfig COMPACTION 235e9e96b39SMel Gorman bool "Allow for memory compaction" 23605106e6aSRik van Riel def_bool y 237e9e96b39SMel Gorman select MIGRATION 23833a93877SAndrea Arcangeli depends on MMU 239e9e96b39SMel Gorman help 240e9e96b39SMel Gorman Allows the compaction of memory for the allocation of huge pages. 241e9e96b39SMel Gorman 242e9e96b39SMel Gorman# 2437cbe34cfSChristoph Lameter# support for page migration 2447cbe34cfSChristoph Lameter# 2457cbe34cfSChristoph Lameterconfig MIGRATION 246b20a3503SChristoph Lameter bool "Page migration" 2476c5240aeSChristoph Lameter def_bool y 24847118af0SMichal Nazarewicz depends on NUMA || ARCH_ENABLE_MEMORY_HOTREMOVE || COMPACTION || CMA 249b20a3503SChristoph Lameter help 250b20a3503SChristoph Lameter Allows the migration of the physical location of pages of processes 251e9e96b39SMel Gorman while the virtual addresses are not changed. This is useful in 252e9e96b39SMel Gorman two situations. The first is on NUMA systems to put pages nearer 253e9e96b39SMel Gorman to the processors accessing. The second is when allocating huge 254e9e96b39SMel Gorman pages as migration can relocate pages to satisfy a huge page 255e9e96b39SMel Gorman allocation instead of reclaiming. 2566550e07fSGreg Kroah-Hartman 257600715dcSJeremy Fitzhardingeconfig PHYS_ADDR_T_64BIT 258600715dcSJeremy Fitzhardinge def_bool 64BIT || ARCH_PHYS_ADDR_T_64BIT 259600715dcSJeremy Fitzhardinge 2604b51d669SChristoph Lameterconfig ZONE_DMA_FLAG 2614b51d669SChristoph Lameter int 2624b51d669SChristoph Lameter default "0" if !ZONE_DMA 2634b51d669SChristoph Lameter default "1" 2644b51d669SChristoph Lameter 2652a7326b5SChristoph Lameterconfig BOUNCE 2669ca24e2eSVinayak Menon bool "Enable bounce buffers" 2679ca24e2eSVinayak Menon default y 2682a7326b5SChristoph Lameter depends on BLOCK && MMU && (ZONE_DMA || HIGHMEM) 2699ca24e2eSVinayak Menon help 2709ca24e2eSVinayak Menon Enable bounce buffers for devices that cannot access 2719ca24e2eSVinayak Menon the full range of memory available to the CPU. Enabled 2729ca24e2eSVinayak Menon by default when ZONE_DMA or HIGHMEM is selected, but you 2739ca24e2eSVinayak Menon may say n to override this. 2742a7326b5SChristoph Lameter 275ffecfd1aSDarrick J. Wong# On the 'tile' arch, USB OHCI needs the bounce pool since tilegx will often 276ffecfd1aSDarrick J. Wong# have more than 4GB of memory, but we don't currently use the IOTLB to present 277ffecfd1aSDarrick J. Wong# a 32-bit address to OHCI. So we need to use a bounce pool instead. 278ffecfd1aSDarrick J. Wong# 279ffecfd1aSDarrick J. Wong# We also use the bounce pool to provide stable page writes for jbd. jbd 280ffecfd1aSDarrick J. Wong# initiates buffer writeback without locking the page or setting PG_writeback, 281ffecfd1aSDarrick J. Wong# and fixing that behavior (a second time; jbd2 doesn't have this problem) is 282ffecfd1aSDarrick J. Wong# a major rework effort. Instead, use the bounce buffer to snapshot pages 283ffecfd1aSDarrick J. Wong# (until jbd goes away). The only jbd user is ext3. 284ffecfd1aSDarrick J. Wongconfig NEED_BOUNCE_POOL 285ffecfd1aSDarrick J. Wong bool 286ffecfd1aSDarrick J. Wong default y if (TILE && USB_OHCI_HCD) || (BLK_DEV_INTEGRITY && JBD) 287ffecfd1aSDarrick J. Wong 2886225e937SChristoph Lameterconfig NR_QUICK 2896225e937SChristoph Lameter int 2906225e937SChristoph Lameter depends on QUICKLIST 2910176bd3dSPaul Mundt default "2" if AVR32 2926225e937SChristoph Lameter default "1" 293f057eac0SStephen Rothwell 294f057eac0SStephen Rothwellconfig VIRT_TO_BUS 2954febd95aSStephen Rothwell bool 2964febd95aSStephen Rothwell help 2974febd95aSStephen Rothwell An architecture should select this if it implements the 2984febd95aSStephen Rothwell deprecated interface virt_to_bus(). All new architectures 2994febd95aSStephen Rothwell should probably not select this. 3004febd95aSStephen Rothwell 301cddb8a5cSAndrea Arcangeli 302cddb8a5cSAndrea Arcangeliconfig MMU_NOTIFIER 303cddb8a5cSAndrea Arcangeli bool 304fc4d5c29SDavid Howells 305f8af4da3SHugh Dickinsconfig KSM 306f8af4da3SHugh Dickins bool "Enable KSM for page merging" 307f8af4da3SHugh Dickins depends on MMU 308f8af4da3SHugh Dickins help 309f8af4da3SHugh Dickins Enable Kernel Samepage Merging: KSM periodically scans those areas 310f8af4da3SHugh Dickins of an application's address space that an app has advised may be 311f8af4da3SHugh Dickins mergeable. When it finds pages of identical content, it replaces 312d0f209f6SHugh Dickins the many instances by a single page with that content, so 313f8af4da3SHugh Dickins saving memory until one or another app needs to modify the content. 314f8af4da3SHugh Dickins Recommended for use with KVM, or with other duplicative applications. 315c73602adSHugh Dickins See Documentation/vm/ksm.txt for more information: KSM is inactive 316c73602adSHugh Dickins until a program has madvised that an area is MADV_MERGEABLE, and 317c73602adSHugh Dickins root has set /sys/kernel/mm/ksm/run to 1 (if CONFIG_SYSFS is set). 318f8af4da3SHugh Dickins 319e0a94c2aSChristoph Lameterconfig DEFAULT_MMAP_MIN_ADDR 320e0a94c2aSChristoph Lameter int "Low address space to protect from user allocation" 3216e141546SDavid Howells depends on MMU 322e0a94c2aSChristoph Lameter default 4096 323e0a94c2aSChristoph Lameter help 324e0a94c2aSChristoph Lameter This is the portion of low virtual memory which should be protected 325e0a94c2aSChristoph Lameter from userspace allocation. Keeping a user from writing to low pages 326e0a94c2aSChristoph Lameter can help reduce the impact of kernel NULL pointer bugs. 327e0a94c2aSChristoph Lameter 328e0a94c2aSChristoph Lameter For most ia64, ppc64 and x86 users with lots of address space 329e0a94c2aSChristoph Lameter a value of 65536 is reasonable and should cause no problems. 330e0a94c2aSChristoph Lameter On arm and other archs it should not be higher than 32768. 331788084abSEric Paris Programs which use vm86 functionality or have some need to map 332788084abSEric Paris this low address space will need CAP_SYS_RAWIO or disable this 333788084abSEric Paris protection by setting the value to 0. 334e0a94c2aSChristoph Lameter 335e0a94c2aSChristoph Lameter This value can be changed after boot using the 336e0a94c2aSChristoph Lameter /proc/sys/vm/mmap_min_addr tunable. 337e0a94c2aSChristoph Lameter 338d949f36fSLinus Torvaldsconfig ARCH_SUPPORTS_MEMORY_FAILURE 339d949f36fSLinus Torvalds bool 340e0a94c2aSChristoph Lameter 3416a46079cSAndi Kleenconfig MEMORY_FAILURE 3426a46079cSAndi Kleen depends on MMU 343d949f36fSLinus Torvalds depends on ARCH_SUPPORTS_MEMORY_FAILURE 3446a46079cSAndi Kleen bool "Enable recovery from hardware memory errors" 345ee6f509cSMinchan Kim select MEMORY_ISOLATION 3466a46079cSAndi Kleen help 3476a46079cSAndi Kleen Enables code to recover from some memory failures on systems 3486a46079cSAndi Kleen with MCA recovery. This allows a system to continue running 3496a46079cSAndi Kleen even when some of its memory has uncorrected errors. This requires 3506a46079cSAndi Kleen special hardware support and typically ECC memory. 3516a46079cSAndi Kleen 352cae681fcSAndi Kleenconfig HWPOISON_INJECT 353413f9efbSAndi Kleen tristate "HWPoison pages injector" 35427df5068SAndi Kleen depends on MEMORY_FAILURE && DEBUG_KERNEL && PROC_FS 355478c5ffcSWu Fengguang select PROC_PAGE_MONITOR 356cae681fcSAndi Kleen 357fc4d5c29SDavid Howellsconfig NOMMU_INITIAL_TRIM_EXCESS 358fc4d5c29SDavid Howells int "Turn on mmap() excess space trimming before booting" 359fc4d5c29SDavid Howells depends on !MMU 360fc4d5c29SDavid Howells default 1 361fc4d5c29SDavid Howells help 362fc4d5c29SDavid Howells The NOMMU mmap() frequently needs to allocate large contiguous chunks 363fc4d5c29SDavid Howells of memory on which to store mappings, but it can only ask the system 364fc4d5c29SDavid Howells allocator for chunks in 2^N*PAGE_SIZE amounts - which is frequently 365fc4d5c29SDavid Howells more than it requires. To deal with this, mmap() is able to trim off 366fc4d5c29SDavid Howells the excess and return it to the allocator. 367fc4d5c29SDavid Howells 368fc4d5c29SDavid Howells If trimming is enabled, the excess is trimmed off and returned to the 369fc4d5c29SDavid Howells system allocator, which can cause extra fragmentation, particularly 370fc4d5c29SDavid Howells if there are a lot of transient processes. 371fc4d5c29SDavid Howells 372fc4d5c29SDavid Howells If trimming is disabled, the excess is kept, but not used, which for 373fc4d5c29SDavid Howells long-term mappings means that the space is wasted. 374fc4d5c29SDavid Howells 375fc4d5c29SDavid Howells Trimming can be dynamically controlled through a sysctl option 376fc4d5c29SDavid Howells (/proc/sys/vm/nr_trim_pages) which specifies the minimum number of 377fc4d5c29SDavid Howells excess pages there must be before trimming should occur, or zero if 378fc4d5c29SDavid Howells no trimming is to occur. 379fc4d5c29SDavid Howells 380fc4d5c29SDavid Howells This option specifies the initial value of this option. The default 381fc4d5c29SDavid Howells of 1 says that all excess pages should be trimmed. 382fc4d5c29SDavid Howells 383fc4d5c29SDavid Howells See Documentation/nommu-mmap.txt for more information. 384bbddff05STejun Heo 3854c76d9d1SAndrea Arcangeliconfig TRANSPARENT_HUGEPAGE 38613ece886SAndrea Arcangeli bool "Transparent Hugepage Support" 38715626062SGerald Schaefer depends on HAVE_ARCH_TRANSPARENT_HUGEPAGE 3885d689240SAndrea Arcangeli select COMPACTION 3894c76d9d1SAndrea Arcangeli help 3904c76d9d1SAndrea Arcangeli Transparent Hugepages allows the kernel to use huge pages and 3914c76d9d1SAndrea Arcangeli huge tlb transparently to the applications whenever possible. 3924c76d9d1SAndrea Arcangeli This feature can improve computing performance to certain 3934c76d9d1SAndrea Arcangeli applications by speeding up page faults during memory 3944c76d9d1SAndrea Arcangeli allocation, by reducing the number of tlb misses and by speeding 3954c76d9d1SAndrea Arcangeli up the pagetable walking. 3964c76d9d1SAndrea Arcangeli 3974c76d9d1SAndrea Arcangeli If memory constrained on embedded, you may want to say N. 3984c76d9d1SAndrea Arcangeli 39913ece886SAndrea Arcangelichoice 40013ece886SAndrea Arcangeli prompt "Transparent Hugepage Support sysfs defaults" 40113ece886SAndrea Arcangeli depends on TRANSPARENT_HUGEPAGE 40213ece886SAndrea Arcangeli default TRANSPARENT_HUGEPAGE_ALWAYS 40313ece886SAndrea Arcangeli help 40413ece886SAndrea Arcangeli Selects the sysfs defaults for Transparent Hugepage Support. 40513ece886SAndrea Arcangeli 40613ece886SAndrea Arcangeli config TRANSPARENT_HUGEPAGE_ALWAYS 40713ece886SAndrea Arcangeli bool "always" 40813ece886SAndrea Arcangeli help 40913ece886SAndrea Arcangeli Enabling Transparent Hugepage always, can increase the 41013ece886SAndrea Arcangeli memory footprint of applications without a guaranteed 41113ece886SAndrea Arcangeli benefit but it will work automatically for all applications. 41213ece886SAndrea Arcangeli 41313ece886SAndrea Arcangeli config TRANSPARENT_HUGEPAGE_MADVISE 41413ece886SAndrea Arcangeli bool "madvise" 41513ece886SAndrea Arcangeli help 41613ece886SAndrea Arcangeli Enabling Transparent Hugepage madvise, will only provide a 41713ece886SAndrea Arcangeli performance improvement benefit to the applications using 41813ece886SAndrea Arcangeli madvise(MADV_HUGEPAGE) but it won't risk to increase the 41913ece886SAndrea Arcangeli memory footprint of applications without a guaranteed 42013ece886SAndrea Arcangeli benefit. 42113ece886SAndrea Arcangeliendchoice 42213ece886SAndrea Arcangeli 4235febcbe9SChristopher Yeohconfig CROSS_MEMORY_ATTACH 4245febcbe9SChristopher Yeoh bool "Cross Memory Support" 4255febcbe9SChristopher Yeoh depends on MMU 4265febcbe9SChristopher Yeoh default y 4275febcbe9SChristopher Yeoh help 4285febcbe9SChristopher Yeoh Enabling this option adds the system calls process_vm_readv and 4295febcbe9SChristopher Yeoh process_vm_writev which allow a process with the correct privileges 4305febcbe9SChristopher Yeoh to directly read from or write to to another process's address space. 4315febcbe9SChristopher Yeoh See the man page for more details. 4325febcbe9SChristopher Yeoh 433bbddff05STejun Heo# 434bbddff05STejun Heo# UP and nommu archs use km based percpu allocator 435bbddff05STejun Heo# 436bbddff05STejun Heoconfig NEED_PER_CPU_KM 437bbddff05STejun Heo depends on !SMP 438bbddff05STejun Heo bool 439bbddff05STejun Heo default y 440077b1f83SDan Magenheimer 441077b1f83SDan Magenheimerconfig CLEANCACHE 442077b1f83SDan Magenheimer bool "Enable cleancache driver to cache clean pages if tmem is present" 443077b1f83SDan Magenheimer default n 444077b1f83SDan Magenheimer help 445077b1f83SDan Magenheimer Cleancache can be thought of as a page-granularity victim cache 446077b1f83SDan Magenheimer for clean pages that the kernel's pageframe replacement algorithm 447077b1f83SDan Magenheimer (PFRA) would like to keep around, but can't since there isn't enough 448077b1f83SDan Magenheimer memory. So when the PFRA "evicts" a page, it first attempts to use 449140a1ef2SMichael Witten cleancache code to put the data contained in that page into 450077b1f83SDan Magenheimer "transcendent memory", memory that is not directly accessible or 451077b1f83SDan Magenheimer addressable by the kernel and is of unknown and possibly 452077b1f83SDan Magenheimer time-varying size. And when a cleancache-enabled 453077b1f83SDan Magenheimer filesystem wishes to access a page in a file on disk, it first 454077b1f83SDan Magenheimer checks cleancache to see if it already contains it; if it does, 455077b1f83SDan Magenheimer the page is copied into the kernel and a disk access is avoided. 456077b1f83SDan Magenheimer When a transcendent memory driver is available (such as zcache or 457077b1f83SDan Magenheimer Xen transcendent memory), a significant I/O reduction 458077b1f83SDan Magenheimer may be achieved. When none is available, all cleancache calls 459077b1f83SDan Magenheimer are reduced to a single pointer-compare-against-NULL resulting 460077b1f83SDan Magenheimer in a negligible performance hit. 461077b1f83SDan Magenheimer 462077b1f83SDan Magenheimer If unsure, say Y to enable cleancache 46327c6aec2SDan Magenheimer 46427c6aec2SDan Magenheimerconfig FRONTSWAP 46527c6aec2SDan Magenheimer bool "Enable frontswap to cache swap pages if tmem is present" 46627c6aec2SDan Magenheimer depends on SWAP 46727c6aec2SDan Magenheimer default n 46827c6aec2SDan Magenheimer help 46927c6aec2SDan Magenheimer Frontswap is so named because it can be thought of as the opposite 47027c6aec2SDan Magenheimer of a "backing" store for a swap device. The data is stored into 47127c6aec2SDan Magenheimer "transcendent memory", memory that is not directly accessible or 47227c6aec2SDan Magenheimer addressable by the kernel and is of unknown and possibly 47327c6aec2SDan Magenheimer time-varying size. When space in transcendent memory is available, 47427c6aec2SDan Magenheimer a significant swap I/O reduction may be achieved. When none is 47527c6aec2SDan Magenheimer available, all frontswap calls are reduced to a single pointer- 47627c6aec2SDan Magenheimer compare-against-NULL resulting in a negligible performance hit 47727c6aec2SDan Magenheimer and swap data is stored as normal on the matching swap device. 47827c6aec2SDan Magenheimer 47927c6aec2SDan Magenheimer If unsure, say Y to enable frontswap. 480f825c736SAneesh Kumar K.V 481f825c736SAneesh Kumar K.Vconfig CMA 482f825c736SAneesh Kumar K.V bool "Contiguous Memory Allocator" 483f825c736SAneesh Kumar K.V depends on HAVE_MEMBLOCK 484f825c736SAneesh Kumar K.V select MIGRATION 485f825c736SAneesh Kumar K.V select MEMORY_ISOLATION 486f825c736SAneesh Kumar K.V help 487f825c736SAneesh Kumar K.V This enables the Contiguous Memory Allocator which allows other 488f825c736SAneesh Kumar K.V subsystems to allocate big physically-contiguous blocks of memory. 489f825c736SAneesh Kumar K.V CMA reserves a region of memory and allows only movable pages to 490f825c736SAneesh Kumar K.V be allocated from it. This way, the kernel can use the memory for 491f825c736SAneesh Kumar K.V pagecache and when a subsystem requests for contiguous area, the 492f825c736SAneesh Kumar K.V allocated pages are migrated away to serve the contiguous request. 493f825c736SAneesh Kumar K.V 494f825c736SAneesh Kumar K.V If unsure, say "n". 495f825c736SAneesh Kumar K.V 496f825c736SAneesh Kumar K.Vconfig CMA_DEBUG 497f825c736SAneesh Kumar K.V bool "CMA debug messages (DEVELOPMENT)" 498f825c736SAneesh Kumar K.V depends on DEBUG_KERNEL && CMA 499f825c736SAneesh Kumar K.V help 500f825c736SAneesh Kumar K.V Turns on debug messages in CMA. This produces KERN_DEBUG 501f825c736SAneesh Kumar K.V messages for every CMA call as well as various messages while 502f825c736SAneesh Kumar K.V processing calls such as dma_alloc_from_contiguous(). 503f825c736SAneesh Kumar K.V This option does not affect warning and error messages. 504bf550fc9SAlexander Graf 5054e2e2770SSeth Jenningsconfig ZBUD 5064e2e2770SSeth Jennings tristate 5074e2e2770SSeth Jennings default n 5084e2e2770SSeth Jennings help 5094e2e2770SSeth Jennings A special purpose allocator for storing compressed pages. 5104e2e2770SSeth Jennings It is designed to store up to two compressed pages per physical 5114e2e2770SSeth Jennings page. While this design limits storage density, it has simple and 5124e2e2770SSeth Jennings deterministic reclaim properties that make it preferable to a higher 5134e2e2770SSeth Jennings density approach when reclaim will be used. 5144e2e2770SSeth Jennings 5152b281117SSeth Jenningsconfig ZSWAP 5162b281117SSeth Jennings bool "Compressed cache for swap pages (EXPERIMENTAL)" 5172b281117SSeth Jennings depends on FRONTSWAP && CRYPTO=y 5182b281117SSeth Jennings select CRYPTO_LZO 5192b281117SSeth Jennings select ZBUD 5202b281117SSeth Jennings default n 5212b281117SSeth Jennings help 5222b281117SSeth Jennings A lightweight compressed cache for swap pages. It takes 5232b281117SSeth Jennings pages that are in the process of being swapped out and attempts to 5242b281117SSeth Jennings compress them into a dynamically allocated RAM-based memory pool. 5252b281117SSeth Jennings This can result in a significant I/O reduction on swap device and, 5262b281117SSeth Jennings in the case where decompressing from RAM is faster that swap device 5272b281117SSeth Jennings reads, can also improve workload performance. 5282b281117SSeth Jennings 5292b281117SSeth Jennings This is marked experimental because it is a new feature (as of 5302b281117SSeth Jennings v3.11) that interacts heavily with memory reclaim. While these 5312b281117SSeth Jennings interactions don't cause any known issues on simple memory setups, 5322b281117SSeth Jennings they have not be fully explored on the large set of potential 5332b281117SSeth Jennings configurations and workloads that exist. 5342b281117SSeth Jennings 5350f8975ecSPavel Emelyanovconfig MEM_SOFT_DIRTY 5360f8975ecSPavel Emelyanov bool "Track memory changes" 5370f8975ecSPavel Emelyanov depends on CHECKPOINT_RESTORE && HAVE_ARCH_SOFT_DIRTY 5380f8975ecSPavel Emelyanov select PROC_PAGE_MONITOR 5390f8975ecSPavel Emelyanov help 5400f8975ecSPavel Emelyanov This option enables memory changes tracking by introducing a 5410f8975ecSPavel Emelyanov soft-dirty bit on pte-s. This bit it set when someone writes 5420f8975ecSPavel Emelyanov into a page just as regular dirty bit, but unlike the latter 5430f8975ecSPavel Emelyanov it can be cleared by hands. 5440f8975ecSPavel Emelyanov 5450f8975ecSPavel Emelyanov See Documentation/vm/soft-dirty.txt for more details. 546