1e1785e85SDave Hansenconfig SELECT_MEMORY_MODEL 2e1785e85SDave Hansen def_bool y 3e1785e85SDave Hansen depends on EXPERIMENTAL || 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. 23d41dee36SAndy Whitcroft DISCONTIGMEM is an 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 1433947be19SDave Hansen# eventually, we can have this option just 'select SPARSEMEM' 1443947be19SDave Hansenconfig MEMORY_HOTPLUG 1453947be19SDave Hansen bool "Allow for memory hot-add" 146ec69acbbSKeith Mannthey depends on SPARSEMEM || X86_64_ACPI_NUMA 1476ad696d2SAndi Kleen depends on HOTPLUG && ARCH_ENABLE_MEMORY_HOTPLUG 148ed84a07aSKumar Gala depends on (IA64 || X86 || PPC_BOOK3S_64 || SUPERH || S390) 1493947be19SDave Hansen 150ec69acbbSKeith Manntheyconfig MEMORY_HOTPLUG_SPARSE 151ec69acbbSKeith Mannthey def_bool y 152ec69acbbSKeith Mannthey depends on SPARSEMEM && MEMORY_HOTPLUG 153ec69acbbSKeith Mannthey 1540c0e6195SKAMEZAWA Hiroyukiconfig MEMORY_HOTREMOVE 1550c0e6195SKAMEZAWA Hiroyuki bool "Allow for memory hot remove" 1560c0e6195SKAMEZAWA Hiroyuki depends on MEMORY_HOTPLUG && ARCH_ENABLE_MEMORY_HOTREMOVE 1570c0e6195SKAMEZAWA Hiroyuki depends on MIGRATION 1580c0e6195SKAMEZAWA Hiroyuki 159e20b8ccaSChristoph Lameter# 160e20b8ccaSChristoph Lameter# If we have space for more page flags then we can enable additional 161e20b8ccaSChristoph Lameter# optimizations and functionality. 162e20b8ccaSChristoph Lameter# 163e20b8ccaSChristoph Lameter# Regular Sparsemem takes page flag bits for the sectionid if it does not 164e20b8ccaSChristoph Lameter# use a virtual memmap. Disable extended page flags for 32 bit platforms 165e20b8ccaSChristoph Lameter# that require the use of a sectionid in the page flags. 166e20b8ccaSChristoph Lameter# 167e20b8ccaSChristoph Lameterconfig PAGEFLAGS_EXTENDED 168e20b8ccaSChristoph Lameter def_bool y 169a269cca9SH. Peter Anvin depends on 64BIT || SPARSEMEM_VMEMMAP || !SPARSEMEM 170e20b8ccaSChristoph Lameter 1714c21e2f2SHugh Dickins# Heavily threaded applications may benefit from splitting the mm-wide 1724c21e2f2SHugh Dickins# page_table_lock, so that faults on different parts of the user address 1734c21e2f2SHugh Dickins# space can be handled with less contention: split it at this NR_CPUS. 1744c21e2f2SHugh Dickins# Default to 4 for wider testing, though 8 might be more appropriate. 1754c21e2f2SHugh Dickins# ARM's adjust_pte (unused if VIPT) depends on mm-wide page_table_lock. 1767b6ac9dfSHugh Dickins# PA-RISC 7xxx's spinlock_t would enlarge struct page from 32 to 44 bytes. 177a70caa8bSHugh Dickins# DEBUG_SPINLOCK and DEBUG_LOCK_ALLOC spinlock_t also enlarge struct page. 1784c21e2f2SHugh Dickins# 1794c21e2f2SHugh Dickinsconfig SPLIT_PTLOCK_CPUS 1804c21e2f2SHugh Dickins int 181a70caa8bSHugh Dickins default "999999" if ARM && !CPU_CACHE_VIPT 182a70caa8bSHugh Dickins default "999999" if PARISC && !PA20 183a70caa8bSHugh Dickins default "999999" if DEBUG_SPINLOCK || DEBUG_LOCK_ALLOC 1844c21e2f2SHugh Dickins default "4" 1857cbe34cfSChristoph Lameter 1867cbe34cfSChristoph Lameter# 187e9e96b39SMel Gorman# support for memory compaction 188e9e96b39SMel Gormanconfig COMPACTION 189e9e96b39SMel Gorman bool "Allow for memory compaction" 190e9e96b39SMel Gorman select MIGRATION 19133a93877SAndrea Arcangeli depends on MMU 192e9e96b39SMel Gorman help 193e9e96b39SMel Gorman Allows the compaction of memory for the allocation of huge pages. 194e9e96b39SMel Gorman 195e9e96b39SMel Gorman# 1967cbe34cfSChristoph Lameter# support for page migration 1977cbe34cfSChristoph Lameter# 1987cbe34cfSChristoph Lameterconfig MIGRATION 199b20a3503SChristoph Lameter bool "Page migration" 2006c5240aeSChristoph Lameter def_bool y 20147118af0SMichal Nazarewicz depends on NUMA || ARCH_ENABLE_MEMORY_HOTREMOVE || COMPACTION || CMA 202b20a3503SChristoph Lameter help 203b20a3503SChristoph Lameter Allows the migration of the physical location of pages of processes 204e9e96b39SMel Gorman while the virtual addresses are not changed. This is useful in 205e9e96b39SMel Gorman two situations. The first is on NUMA systems to put pages nearer 206e9e96b39SMel Gorman to the processors accessing. The second is when allocating huge 207e9e96b39SMel Gorman pages as migration can relocate pages to satisfy a huge page 208e9e96b39SMel Gorman allocation instead of reclaiming. 2096550e07fSGreg Kroah-Hartman 210600715dcSJeremy Fitzhardingeconfig PHYS_ADDR_T_64BIT 211600715dcSJeremy Fitzhardinge def_bool 64BIT || ARCH_PHYS_ADDR_T_64BIT 212600715dcSJeremy Fitzhardinge 2134b51d669SChristoph Lameterconfig ZONE_DMA_FLAG 2144b51d669SChristoph Lameter int 2154b51d669SChristoph Lameter default "0" if !ZONE_DMA 2164b51d669SChristoph Lameter default "1" 2174b51d669SChristoph Lameter 2182a7326b5SChristoph Lameterconfig BOUNCE 2192a7326b5SChristoph Lameter def_bool y 2202a7326b5SChristoph Lameter depends on BLOCK && MMU && (ZONE_DMA || HIGHMEM) 2212a7326b5SChristoph Lameter 2226225e937SChristoph Lameterconfig NR_QUICK 2236225e937SChristoph Lameter int 2246225e937SChristoph Lameter depends on QUICKLIST 2250176bd3dSPaul Mundt default "2" if AVR32 2266225e937SChristoph Lameter default "1" 227f057eac0SStephen Rothwell 228f057eac0SStephen Rothwellconfig VIRT_TO_BUS 229f057eac0SStephen Rothwell def_bool y 230f057eac0SStephen Rothwell depends on !ARCH_NO_VIRT_TO_BUS 231cddb8a5cSAndrea Arcangeli 232cddb8a5cSAndrea Arcangeliconfig MMU_NOTIFIER 233cddb8a5cSAndrea Arcangeli bool 234fc4d5c29SDavid Howells 235f8af4da3SHugh Dickinsconfig KSM 236f8af4da3SHugh Dickins bool "Enable KSM for page merging" 237f8af4da3SHugh Dickins depends on MMU 238f8af4da3SHugh Dickins help 239f8af4da3SHugh Dickins Enable Kernel Samepage Merging: KSM periodically scans those areas 240f8af4da3SHugh Dickins of an application's address space that an app has advised may be 241f8af4da3SHugh Dickins mergeable. When it finds pages of identical content, it replaces 242d0f209f6SHugh Dickins the many instances by a single page with that content, so 243f8af4da3SHugh Dickins saving memory until one or another app needs to modify the content. 244f8af4da3SHugh Dickins Recommended for use with KVM, or with other duplicative applications. 245c73602adSHugh Dickins See Documentation/vm/ksm.txt for more information: KSM is inactive 246c73602adSHugh Dickins until a program has madvised that an area is MADV_MERGEABLE, and 247c73602adSHugh Dickins root has set /sys/kernel/mm/ksm/run to 1 (if CONFIG_SYSFS is set). 248f8af4da3SHugh Dickins 249e0a94c2aSChristoph Lameterconfig DEFAULT_MMAP_MIN_ADDR 250e0a94c2aSChristoph Lameter int "Low address space to protect from user allocation" 2516e141546SDavid Howells depends on MMU 252e0a94c2aSChristoph Lameter default 4096 253e0a94c2aSChristoph Lameter help 254e0a94c2aSChristoph Lameter This is the portion of low virtual memory which should be protected 255e0a94c2aSChristoph Lameter from userspace allocation. Keeping a user from writing to low pages 256e0a94c2aSChristoph Lameter can help reduce the impact of kernel NULL pointer bugs. 257e0a94c2aSChristoph Lameter 258e0a94c2aSChristoph Lameter For most ia64, ppc64 and x86 users with lots of address space 259e0a94c2aSChristoph Lameter a value of 65536 is reasonable and should cause no problems. 260e0a94c2aSChristoph Lameter On arm and other archs it should not be higher than 32768. 261788084abSEric Paris Programs which use vm86 functionality or have some need to map 262788084abSEric Paris this low address space will need CAP_SYS_RAWIO or disable this 263788084abSEric Paris protection by setting the value to 0. 264e0a94c2aSChristoph Lameter 265e0a94c2aSChristoph Lameter This value can be changed after boot using the 266e0a94c2aSChristoph Lameter /proc/sys/vm/mmap_min_addr tunable. 267e0a94c2aSChristoph Lameter 268d949f36fSLinus Torvaldsconfig ARCH_SUPPORTS_MEMORY_FAILURE 269d949f36fSLinus Torvalds bool 270e0a94c2aSChristoph Lameter 2716a46079cSAndi Kleenconfig MEMORY_FAILURE 2726a46079cSAndi Kleen depends on MMU 273d949f36fSLinus Torvalds depends on ARCH_SUPPORTS_MEMORY_FAILURE 2746a46079cSAndi Kleen bool "Enable recovery from hardware memory errors" 2756a46079cSAndi Kleen help 2766a46079cSAndi Kleen Enables code to recover from some memory failures on systems 2776a46079cSAndi Kleen with MCA recovery. This allows a system to continue running 2786a46079cSAndi Kleen even when some of its memory has uncorrected errors. This requires 2796a46079cSAndi Kleen special hardware support and typically ECC memory. 2806a46079cSAndi Kleen 281cae681fcSAndi Kleenconfig HWPOISON_INJECT 282413f9efbSAndi Kleen tristate "HWPoison pages injector" 28327df5068SAndi Kleen depends on MEMORY_FAILURE && DEBUG_KERNEL && PROC_FS 284478c5ffcSWu Fengguang select PROC_PAGE_MONITOR 285cae681fcSAndi Kleen 286fc4d5c29SDavid Howellsconfig NOMMU_INITIAL_TRIM_EXCESS 287fc4d5c29SDavid Howells int "Turn on mmap() excess space trimming before booting" 288fc4d5c29SDavid Howells depends on !MMU 289fc4d5c29SDavid Howells default 1 290fc4d5c29SDavid Howells help 291fc4d5c29SDavid Howells The NOMMU mmap() frequently needs to allocate large contiguous chunks 292fc4d5c29SDavid Howells of memory on which to store mappings, but it can only ask the system 293fc4d5c29SDavid Howells allocator for chunks in 2^N*PAGE_SIZE amounts - which is frequently 294fc4d5c29SDavid Howells more than it requires. To deal with this, mmap() is able to trim off 295fc4d5c29SDavid Howells the excess and return it to the allocator. 296fc4d5c29SDavid Howells 297fc4d5c29SDavid Howells If trimming is enabled, the excess is trimmed off and returned to the 298fc4d5c29SDavid Howells system allocator, which can cause extra fragmentation, particularly 299fc4d5c29SDavid Howells if there are a lot of transient processes. 300fc4d5c29SDavid Howells 301fc4d5c29SDavid Howells If trimming is disabled, the excess is kept, but not used, which for 302fc4d5c29SDavid Howells long-term mappings means that the space is wasted. 303fc4d5c29SDavid Howells 304fc4d5c29SDavid Howells Trimming can be dynamically controlled through a sysctl option 305fc4d5c29SDavid Howells (/proc/sys/vm/nr_trim_pages) which specifies the minimum number of 306fc4d5c29SDavid Howells excess pages there must be before trimming should occur, or zero if 307fc4d5c29SDavid Howells no trimming is to occur. 308fc4d5c29SDavid Howells 309fc4d5c29SDavid Howells This option specifies the initial value of this option. The default 310fc4d5c29SDavid Howells of 1 says that all excess pages should be trimmed. 311fc4d5c29SDavid Howells 312fc4d5c29SDavid Howells See Documentation/nommu-mmap.txt for more information. 313bbddff05STejun Heo 3144c76d9d1SAndrea Arcangeliconfig TRANSPARENT_HUGEPAGE 31513ece886SAndrea Arcangeli bool "Transparent Hugepage Support" 316f2d6bfe9SJohannes Weiner depends on X86 && MMU 3175d689240SAndrea Arcangeli select COMPACTION 3184c76d9d1SAndrea Arcangeli help 3194c76d9d1SAndrea Arcangeli Transparent Hugepages allows the kernel to use huge pages and 3204c76d9d1SAndrea Arcangeli huge tlb transparently to the applications whenever possible. 3214c76d9d1SAndrea Arcangeli This feature can improve computing performance to certain 3224c76d9d1SAndrea Arcangeli applications by speeding up page faults during memory 3234c76d9d1SAndrea Arcangeli allocation, by reducing the number of tlb misses and by speeding 3244c76d9d1SAndrea Arcangeli up the pagetable walking. 3254c76d9d1SAndrea Arcangeli 3264c76d9d1SAndrea Arcangeli If memory constrained on embedded, you may want to say N. 3274c76d9d1SAndrea Arcangeli 32813ece886SAndrea Arcangelichoice 32913ece886SAndrea Arcangeli prompt "Transparent Hugepage Support sysfs defaults" 33013ece886SAndrea Arcangeli depends on TRANSPARENT_HUGEPAGE 33113ece886SAndrea Arcangeli default TRANSPARENT_HUGEPAGE_ALWAYS 33213ece886SAndrea Arcangeli help 33313ece886SAndrea Arcangeli Selects the sysfs defaults for Transparent Hugepage Support. 33413ece886SAndrea Arcangeli 33513ece886SAndrea Arcangeli config TRANSPARENT_HUGEPAGE_ALWAYS 33613ece886SAndrea Arcangeli bool "always" 33713ece886SAndrea Arcangeli help 33813ece886SAndrea Arcangeli Enabling Transparent Hugepage always, can increase the 33913ece886SAndrea Arcangeli memory footprint of applications without a guaranteed 34013ece886SAndrea Arcangeli benefit but it will work automatically for all applications. 34113ece886SAndrea Arcangeli 34213ece886SAndrea Arcangeli config TRANSPARENT_HUGEPAGE_MADVISE 34313ece886SAndrea Arcangeli bool "madvise" 34413ece886SAndrea Arcangeli help 34513ece886SAndrea Arcangeli Enabling Transparent Hugepage madvise, will only provide a 34613ece886SAndrea Arcangeli performance improvement benefit to the applications using 34713ece886SAndrea Arcangeli madvise(MADV_HUGEPAGE) but it won't risk to increase the 34813ece886SAndrea Arcangeli memory footprint of applications without a guaranteed 34913ece886SAndrea Arcangeli benefit. 35013ece886SAndrea Arcangeliendchoice 35113ece886SAndrea Arcangeli 352bbddff05STejun Heo# 353bbddff05STejun Heo# UP and nommu archs use km based percpu allocator 354bbddff05STejun Heo# 355bbddff05STejun Heoconfig NEED_PER_CPU_KM 356bbddff05STejun Heo depends on !SMP 357bbddff05STejun Heo bool 358bbddff05STejun Heo default y 359077b1f83SDan Magenheimer 360077b1f83SDan Magenheimerconfig CLEANCACHE 361077b1f83SDan Magenheimer bool "Enable cleancache driver to cache clean pages if tmem is present" 362077b1f83SDan Magenheimer default n 363077b1f83SDan Magenheimer help 364077b1f83SDan Magenheimer Cleancache can be thought of as a page-granularity victim cache 365077b1f83SDan Magenheimer for clean pages that the kernel's pageframe replacement algorithm 366077b1f83SDan Magenheimer (PFRA) would like to keep around, but can't since there isn't enough 367077b1f83SDan Magenheimer memory. So when the PFRA "evicts" a page, it first attempts to use 368140a1ef2SMichael Witten cleancache code to put the data contained in that page into 369077b1f83SDan Magenheimer "transcendent memory", memory that is not directly accessible or 370077b1f83SDan Magenheimer addressable by the kernel and is of unknown and possibly 371077b1f83SDan Magenheimer time-varying size. And when a cleancache-enabled 372077b1f83SDan Magenheimer filesystem wishes to access a page in a file on disk, it first 373077b1f83SDan Magenheimer checks cleancache to see if it already contains it; if it does, 374077b1f83SDan Magenheimer the page is copied into the kernel and a disk access is avoided. 375077b1f83SDan Magenheimer When a transcendent memory driver is available (such as zcache or 376077b1f83SDan Magenheimer Xen transcendent memory), a significant I/O reduction 377077b1f83SDan Magenheimer may be achieved. When none is available, all cleancache calls 378077b1f83SDan Magenheimer are reduced to a single pointer-compare-against-NULL resulting 379077b1f83SDan Magenheimer in a negligible performance hit. 380077b1f83SDan Magenheimer 381077b1f83SDan Magenheimer If unsure, say Y to enable cleancache 382