Based on kernel version 4.16.1. Page generated on 2018-04-09 11:53 EST.
1 Documentation for /proc/sys/vm/* kernel version 2.6.29 2 (c) 1998, 1999, Rik van Riel <riel@nl.linux.org> 3 (c) 2008 Peter W. Morreale <pmorreale@novell.com> 4 5 For general info and legal blurb, please look in README. 6 7 ============================================================== 8 9 This file contains the documentation for the sysctl files in 10 /proc/sys/vm and is valid for Linux kernel version 2.6.29. 11 12 The files in this directory can be used to tune the operation 13 of the virtual memory (VM) subsystem of the Linux kernel and 14 the writeout of dirty data to disk. 15 16 Default values and initialization routines for most of these 17 files can be found in mm/swap.c. 18 19 Currently, these files are in /proc/sys/vm: 20 21 - admin_reserve_kbytes 22 - block_dump 23 - compact_memory 24 - compact_unevictable_allowed 25 - dirty_background_bytes 26 - dirty_background_ratio 27 - dirty_bytes 28 - dirty_expire_centisecs 29 - dirty_ratio 30 - dirty_writeback_centisecs 31 - drop_caches 32 - extfrag_threshold 33 - hugetlb_shm_group 34 - laptop_mode 35 - legacy_va_layout 36 - lowmem_reserve_ratio 37 - max_map_count 38 - memory_failure_early_kill 39 - memory_failure_recovery 40 - min_free_kbytes 41 - min_slab_ratio 42 - min_unmapped_ratio 43 - mmap_min_addr 44 - mmap_rnd_bits 45 - mmap_rnd_compat_bits 46 - nr_hugepages 47 - nr_overcommit_hugepages 48 - nr_trim_pages (only if CONFIG_MMU=n) 49 - numa_zonelist_order 50 - oom_dump_tasks 51 - oom_kill_allocating_task 52 - overcommit_kbytes 53 - overcommit_memory 54 - overcommit_ratio 55 - page-cluster 56 - panic_on_oom 57 - percpu_pagelist_fraction 58 - stat_interval 59 - stat_refresh 60 - numa_stat 61 - swappiness 62 - user_reserve_kbytes 63 - vfs_cache_pressure 64 - watermark_scale_factor 65 - zone_reclaim_mode 66 67 ============================================================== 68 69 admin_reserve_kbytes 70 71 The amount of free memory in the system that should be reserved for users 72 with the capability cap_sys_admin. 73 74 admin_reserve_kbytes defaults to min(3% of free pages, 8MB) 75 76 That should provide enough for the admin to log in and kill a process, 77 if necessary, under the default overcommit 'guess' mode. 78 79 Systems running under overcommit 'never' should increase this to account 80 for the full Virtual Memory Size of programs used to recover. Otherwise, 81 root may not be able to log in to recover the system. 82 83 How do you calculate a minimum useful reserve? 84 85 sshd or login + bash (or some other shell) + top (or ps, kill, etc.) 86 87 For overcommit 'guess', we can sum resident set sizes (RSS). 88 On x86_64 this is about 8MB. 89 90 For overcommit 'never', we can take the max of their virtual sizes (VSZ) 91 and add the sum of their RSS. 92 On x86_64 this is about 128MB. 93 94 Changing this takes effect whenever an application requests memory. 95 96 ============================================================== 97 98 block_dump 99 100 block_dump enables block I/O debugging when set to a nonzero value. More 101 information on block I/O debugging is in Documentation/laptops/laptop-mode.txt. 102 103 ============================================================== 104 105 compact_memory 106 107 Available only when CONFIG_COMPACTION is set. When 1 is written to the file, 108 all zones are compacted such that free memory is available in contiguous 109 blocks where possible. This can be important for example in the allocation of 110 huge pages although processes will also directly compact memory as required. 111 112 ============================================================== 113 114 compact_unevictable_allowed 115 116 Available only when CONFIG_COMPACTION is set. When set to 1, compaction is 117 allowed to examine the unevictable lru (mlocked pages) for pages to compact. 118 This should be used on systems where stalls for minor page faults are an 119 acceptable trade for large contiguous free memory. Set to 0 to prevent 120 compaction from moving pages that are unevictable. Default value is 1. 121 122 ============================================================== 123 124 dirty_background_bytes 125 126 Contains the amount of dirty memory at which the background kernel 127 flusher threads will start writeback. 128 129 Note: dirty_background_bytes is the counterpart of dirty_background_ratio. Only 130 one of them may be specified at a time. When one sysctl is written it is 131 immediately taken into account to evaluate the dirty memory limits and the 132 other appears as 0 when read. 133 134 ============================================================== 135 136 dirty_background_ratio 137 138 Contains, as a percentage of total available memory that contains free pages 139 and reclaimable pages, the number of pages at which the background kernel 140 flusher threads will start writing out dirty data. 141 142 The total available memory is not equal to total system memory. 143 144 ============================================================== 145 146 dirty_bytes 147 148 Contains the amount of dirty memory at which a process generating disk writes 149 will itself start writeback. 150 151 Note: dirty_bytes is the counterpart of dirty_ratio. Only one of them may be 152 specified at a time. When one sysctl is written it is immediately taken into 153 account to evaluate the dirty memory limits and the other appears as 0 when 154 read. 155 156 Note: the minimum value allowed for dirty_bytes is two pages (in bytes); any 157 value lower than this limit will be ignored and the old configuration will be 158 retained. 159 160 ============================================================== 161 162 dirty_expire_centisecs 163 164 This tunable is used to define when dirty data is old enough to be eligible 165 for writeout by the kernel flusher threads. It is expressed in 100'ths 166 of a second. Data which has been dirty in-memory for longer than this 167 interval will be written out next time a flusher thread wakes up. 168 169 ============================================================== 170 171 dirty_ratio 172 173 Contains, as a percentage of total available memory that contains free pages 174 and reclaimable pages, the number of pages at which a process which is 175 generating disk writes will itself start writing out dirty data. 176 177 The total available memory is not equal to total system memory. 178 179 ============================================================== 180 181 dirty_writeback_centisecs 182 183 The kernel flusher threads will periodically wake up and write `old' data 184 out to disk. This tunable expresses the interval between those wakeups, in 185 100'ths of a second. 186 187 Setting this to zero disables periodic writeback altogether. 188 189 ============================================================== 190 191 drop_caches 192 193 Writing to this will cause the kernel to drop clean caches, as well as 194 reclaimable slab objects like dentries and inodes. Once dropped, their 195 memory becomes free. 196 197 To free pagecache: 198 echo 1 > /proc/sys/vm/drop_caches 199 To free reclaimable slab objects (includes dentries and inodes): 200 echo 2 > /proc/sys/vm/drop_caches 201 To free slab objects and pagecache: 202 echo 3 > /proc/sys/vm/drop_caches 203 204 This is a non-destructive operation and will not free any dirty objects. 205 To increase the number of objects freed by this operation, the user may run 206 `sync' prior to writing to /proc/sys/vm/drop_caches. This will minimize the 207 number of dirty objects on the system and create more candidates to be 208 dropped. 209 210 This file is not a means to control the growth of the various kernel caches 211 (inodes, dentries, pagecache, etc...) These objects are automatically 212 reclaimed by the kernel when memory is needed elsewhere on the system. 213 214 Use of this file can cause performance problems. Since it discards cached 215 objects, it may cost a significant amount of I/O and CPU to recreate the 216 dropped objects, especially if they were under heavy use. Because of this, 217 use outside of a testing or debugging environment is not recommended. 218 219 You may see informational messages in your kernel log when this file is 220 used: 221 222 cat (1234): drop_caches: 3 223 224 These are informational only. They do not mean that anything is wrong 225 with your system. To disable them, echo 4 (bit 3) into drop_caches. 226 227 ============================================================== 228 229 extfrag_threshold 230 231 This parameter affects whether the kernel will compact memory or direct 232 reclaim to satisfy a high-order allocation. The extfrag/extfrag_index file in 233 debugfs shows what the fragmentation index for each order is in each zone in 234 the system. Values tending towards 0 imply allocations would fail due to lack 235 of memory, values towards 1000 imply failures are due to fragmentation and -1 236 implies that the allocation will succeed as long as watermarks are met. 237 238 The kernel will not compact memory in a zone if the 239 fragmentation index is <= extfrag_threshold. The default value is 500. 240 241 ============================================================== 242 243 highmem_is_dirtyable 244 245 Available only for systems with CONFIG_HIGHMEM enabled (32b systems). 246 247 This parameter controls whether the high memory is considered for dirty 248 writers throttling. This is not the case by default which means that 249 only the amount of memory directly visible/usable by the kernel can 250 be dirtied. As a result, on systems with a large amount of memory and 251 lowmem basically depleted writers might be throttled too early and 252 streaming writes can get very slow. 253 254 Changing the value to non zero would allow more memory to be dirtied 255 and thus allow writers to write more data which can be flushed to the 256 storage more effectively. Note this also comes with a risk of pre-mature 257 OOM killer because some writers (e.g. direct block device writes) can 258 only use the low memory and they can fill it up with dirty data without 259 any throttling. 260 261 ============================================================== 262 263 hugetlb_shm_group 264 265 hugetlb_shm_group contains group id that is allowed to create SysV 266 shared memory segment using hugetlb page. 267 268 ============================================================== 269 270 laptop_mode 271 272 laptop_mode is a knob that controls "laptop mode". All the things that are 273 controlled by this knob are discussed in Documentation/laptops/laptop-mode.txt. 274 275 ============================================================== 276 277 legacy_va_layout 278 279 If non-zero, this sysctl disables the new 32-bit mmap layout - the kernel 280 will use the legacy (2.4) layout for all processes. 281 282 ============================================================== 283 284 lowmem_reserve_ratio 285 286 For some specialised workloads on highmem machines it is dangerous for 287 the kernel to allow process memory to be allocated from the "lowmem" 288 zone. This is because that memory could then be pinned via the mlock() 289 system call, or by unavailability of swapspace. 290 291 And on large highmem machines this lack of reclaimable lowmem memory 292 can be fatal. 293 294 So the Linux page allocator has a mechanism which prevents allocations 295 which _could_ use highmem from using too much lowmem. This means that 296 a certain amount of lowmem is defended from the possibility of being 297 captured into pinned user memory. 298 299 (The same argument applies to the old 16 megabyte ISA DMA region. This 300 mechanism will also defend that region from allocations which could use 301 highmem or lowmem). 302 303 The `lowmem_reserve_ratio' tunable determines how aggressive the kernel is 304 in defending these lower zones. 305 306 If you have a machine which uses highmem or ISA DMA and your 307 applications are using mlock(), or if you are running with no swap then 308 you probably should change the lowmem_reserve_ratio setting. 309 310 The lowmem_reserve_ratio is an array. You can see them by reading this file. 311 - 312 % cat /proc/sys/vm/lowmem_reserve_ratio 313 256 256 32 314 - 315 Note: # of this elements is one fewer than number of zones. Because the highest 316 zone's value is not necessary for following calculation. 317 318 But, these values are not used directly. The kernel calculates # of protection 319 pages for each zones from them. These are shown as array of protection pages 320 in /proc/zoneinfo like followings. (This is an example of x86-64 box). 321 Each zone has an array of protection pages like this. 322 323 - 324 Node 0, zone DMA 325 pages free 1355 326 min 3 327 low 3 328 high 4 329 : 330 : 331 numa_other 0 332 protection: (0, 2004, 2004, 2004) 333 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 334 pagesets 335 cpu: 0 pcp: 0 336 : 337 - 338 These protections are added to score to judge whether this zone should be used 339 for page allocation or should be reclaimed. 340 341 In this example, if normal pages (index=2) are required to this DMA zone and 342 watermark[WMARK_HIGH] is used for watermark, the kernel judges this zone should 343 not be used because pages_free(1355) is smaller than watermark + protection[2] 344 (4 + 2004 = 2008). If this protection value is 0, this zone would be used for 345 normal page requirement. If requirement is DMA zone(index=0), protection[0] 346 (=0) is used. 347 348 zone[i]'s protection[j] is calculated by following expression. 349 350 (i < j): 351 zone[i]->protection[j] 352 = (total sums of managed_pages from zone[i+1] to zone[j] on the node) 353 / lowmem_reserve_ratio[i]; 354 (i = j): 355 (should not be protected. = 0; 356 (i > j): 357 (not necessary, but looks 0) 358 359 The default values of lowmem_reserve_ratio[i] are 360 256 (if zone[i] means DMA or DMA32 zone) 361 32 (others). 362 As above expression, they are reciprocal number of ratio. 363 256 means 1/256. # of protection pages becomes about "0.39%" of total managed 364 pages of higher zones on the node. 365 366 If you would like to protect more pages, smaller values are effective. 367 The minimum value is 1 (1/1 -> 100%). 368 369 ============================================================== 370 371 max_map_count: 372 373 This file contains the maximum number of memory map areas a process 374 may have. Memory map areas are used as a side-effect of calling 375 malloc, directly by mmap, mprotect, and madvise, and also when loading 376 shared libraries. 377 378 While most applications need less than a thousand maps, certain 379 programs, particularly malloc debuggers, may consume lots of them, 380 e.g., up to one or two maps per allocation. 381 382 The default value is 65536. 383 384 ============================================================= 385 386 memory_failure_early_kill: 387 388 Control how to kill processes when uncorrected memory error (typically 389 a 2bit error in a memory module) is detected in the background by hardware 390 that cannot be handled by the kernel. In some cases (like the page 391 still having a valid copy on disk) the kernel will handle the failure 392 transparently without affecting any applications. But if there is 393 no other uptodate copy of the data it will kill to prevent any data 394 corruptions from propagating. 395 396 1: Kill all processes that have the corrupted and not reloadable page mapped 397 as soon as the corruption is detected. Note this is not supported 398 for a few types of pages, like kernel internally allocated data or 399 the swap cache, but works for the majority of user pages. 400 401 0: Only unmap the corrupted page from all processes and only kill a process 402 who tries to access it. 403 404 The kill is done using a catchable SIGBUS with BUS_MCEERR_AO, so processes can 405 handle this if they want to. 406 407 This is only active on architectures/platforms with advanced machine 408 check handling and depends on the hardware capabilities. 409 410 Applications can override this setting individually with the PR_MCE_KILL prctl 411 412 ============================================================== 413 414 memory_failure_recovery 415 416 Enable memory failure recovery (when supported by the platform) 417 418 1: Attempt recovery. 419 420 0: Always panic on a memory failure. 421 422 ============================================================== 423 424 min_free_kbytes: 425 426 This is used to force the Linux VM to keep a minimum number 427 of kilobytes free. The VM uses this number to compute a 428 watermark[WMARK_MIN] value for each lowmem zone in the system. 429 Each lowmem zone gets a number of reserved free pages based 430 proportionally on its size. 431 432 Some minimal amount of memory is needed to satisfy PF_MEMALLOC 433 allocations; if you set this to lower than 1024KB, your system will 434 become subtly broken, and prone to deadlock under high loads. 435 436 Setting this too high will OOM your machine instantly. 437 438 ============================================================= 439 440 min_slab_ratio: 441 442 This is available only on NUMA kernels. 443 444 A percentage of the total pages in each zone. On Zone reclaim 445 (fallback from the local zone occurs) slabs will be reclaimed if more 446 than this percentage of pages in a zone are reclaimable slab pages. 447 This insures that the slab growth stays under control even in NUMA 448 systems that rarely perform global reclaim. 449 450 The default is 5 percent. 451 452 Note that slab reclaim is triggered in a per zone / node fashion. 453 The process of reclaiming slab memory is currently not node specific 454 and may not be fast. 455 456 ============================================================= 457 458 min_unmapped_ratio: 459 460 This is available only on NUMA kernels. 461 462 This is a percentage of the total pages in each zone. Zone reclaim will 463 only occur if more than this percentage of pages are in a state that 464 zone_reclaim_mode allows to be reclaimed. 465 466 If zone_reclaim_mode has the value 4 OR'd, then the percentage is compared 467 against all file-backed unmapped pages including swapcache pages and tmpfs 468 files. Otherwise, only unmapped pages backed by normal files but not tmpfs 469 files and similar are considered. 470 471 The default is 1 percent. 472 473 ============================================================== 474 475 mmap_min_addr 476 477 This file indicates the amount of address space which a user process will 478 be restricted from mmapping. Since kernel null dereference bugs could 479 accidentally operate based on the information in the first couple of pages 480 of memory userspace processes should not be allowed to write to them. By 481 default this value is set to 0 and no protections will be enforced by the 482 security module. Setting this value to something like 64k will allow the 483 vast majority of applications to work correctly and provide defense in depth 484 against future potential kernel bugs. 485 486 ============================================================== 487 488 mmap_rnd_bits: 489 490 This value can be used to select the number of bits to use to 491 determine the random offset to the base address of vma regions 492 resulting from mmap allocations on architectures which support 493 tuning address space randomization. This value will be bounded 494 by the architecture's minimum and maximum supported values. 495 496 This value can be changed after boot using the 497 /proc/sys/vm/mmap_rnd_bits tunable 498 499 ============================================================== 500 501 mmap_rnd_compat_bits: 502 503 This value can be used to select the number of bits to use to 504 determine the random offset to the base address of vma regions 505 resulting from mmap allocations for applications run in 506 compatibility mode on architectures which support tuning address 507 space randomization. This value will be bounded by the 508 architecture's minimum and maximum supported values. 509 510 This value can be changed after boot using the 511 /proc/sys/vm/mmap_rnd_compat_bits tunable 512 513 ============================================================== 514 515 nr_hugepages 516 517 Change the minimum size of the hugepage pool. 518 519 See Documentation/vm/hugetlbpage.txt 520 521 ============================================================== 522 523 nr_overcommit_hugepages 524 525 Change the maximum size of the hugepage pool. The maximum is 526 nr_hugepages + nr_overcommit_hugepages. 527 528 See Documentation/vm/hugetlbpage.txt 529 530 ============================================================== 531 532 nr_trim_pages 533 534 This is available only on NOMMU kernels. 535 536 This value adjusts the excess page trimming behaviour of power-of-2 aligned 537 NOMMU mmap allocations. 538 539 A value of 0 disables trimming of allocations entirely, while a value of 1 540 trims excess pages aggressively. Any value >= 1 acts as the watermark where 541 trimming of allocations is initiated. 542 543 The default value is 1. 544 545 See Documentation/nommu-mmap.txt for more information. 546 547 ============================================================== 548 549 numa_zonelist_order 550 551 This sysctl is only for NUMA and it is deprecated. Anything but 552 Node order will fail! 553 554 'where the memory is allocated from' is controlled by zonelists. 555 (This documentation ignores ZONE_HIGHMEM/ZONE_DMA32 for simple explanation. 556 you may be able to read ZONE_DMA as ZONE_DMA32...) 557 558 In non-NUMA case, a zonelist for GFP_KERNEL is ordered as following. 559 ZONE_NORMAL -> ZONE_DMA 560 This means that a memory allocation request for GFP_KERNEL will 561 get memory from ZONE_DMA only when ZONE_NORMAL is not available. 562 563 In NUMA case, you can think of following 2 types of order. 564 Assume 2 node NUMA and below is zonelist of Node(0)'s GFP_KERNEL 565 566 (A) Node(0) ZONE_NORMAL -> Node(0) ZONE_DMA -> Node(1) ZONE_NORMAL 567 (B) Node(0) ZONE_NORMAL -> Node(1) ZONE_NORMAL -> Node(0) ZONE_DMA. 568 569 Type(A) offers the best locality for processes on Node(0), but ZONE_DMA 570 will be used before ZONE_NORMAL exhaustion. This increases possibility of 571 out-of-memory(OOM) of ZONE_DMA because ZONE_DMA is tend to be small. 572 573 Type(B) cannot offer the best locality but is more robust against OOM of 574 the DMA zone. 575 576 Type(A) is called as "Node" order. Type (B) is "Zone" order. 577 578 "Node order" orders the zonelists by node, then by zone within each node. 579 Specify "[Nn]ode" for node order 580 581 "Zone Order" orders the zonelists by zone type, then by node within each 582 zone. Specify "[Zz]one" for zone order. 583 584 Specify "[Dd]efault" to request automatic configuration. 585 586 On 32-bit, the Normal zone needs to be preserved for allocations accessible 587 by the kernel, so "zone" order will be selected. 588 589 On 64-bit, devices that require DMA32/DMA are relatively rare, so "node" 590 order will be selected. 591 592 Default order is recommended unless this is causing problems for your 593 system/application. 594 595 ============================================================== 596 597 oom_dump_tasks 598 599 Enables a system-wide task dump (excluding kernel threads) to be produced 600 when the kernel performs an OOM-killing and includes such information as 601 pid, uid, tgid, vm size, rss, pgtables_bytes, swapents, oom_score_adj 602 score, and name. This is helpful to determine why the OOM killer was 603 invoked, to identify the rogue task that caused it, and to determine why 604 the OOM killer chose the task it did to kill. 605 606 If this is set to zero, this information is suppressed. On very 607 large systems with thousands of tasks it may not be feasible to dump 608 the memory state information for each one. Such systems should not 609 be forced to incur a performance penalty in OOM conditions when the 610 information may not be desired. 611 612 If this is set to non-zero, this information is shown whenever the 613 OOM killer actually kills a memory-hogging task. 614 615 The default value is 1 (enabled). 616 617 ============================================================== 618 619 oom_kill_allocating_task 620 621 This enables or disables killing the OOM-triggering task in 622 out-of-memory situations. 623 624 If this is set to zero, the OOM killer will scan through the entire 625 tasklist and select a task based on heuristics to kill. This normally 626 selects a rogue memory-hogging task that frees up a large amount of 627 memory when killed. 628 629 If this is set to non-zero, the OOM killer simply kills the task that 630 triggered the out-of-memory condition. This avoids the expensive 631 tasklist scan. 632 633 If panic_on_oom is selected, it takes precedence over whatever value 634 is used in oom_kill_allocating_task. 635 636 The default value is 0. 637 638 ============================================================== 639 640 overcommit_kbytes: 641 642 When overcommit_memory is set to 2, the committed address space is not 643 permitted to exceed swap plus this amount of physical RAM. See below. 644 645 Note: overcommit_kbytes is the counterpart of overcommit_ratio. Only one 646 of them may be specified at a time. Setting one disables the other (which 647 then appears as 0 when read). 648 649 ============================================================== 650 651 overcommit_memory: 652 653 This value contains a flag that enables memory overcommitment. 654 655 When this flag is 0, the kernel attempts to estimate the amount 656 of free memory left when userspace requests more memory. 657 658 When this flag is 1, the kernel pretends there is always enough 659 memory until it actually runs out. 660 661 When this flag is 2, the kernel uses a "never overcommit" 662 policy that attempts to prevent any overcommit of memory. 663 Note that user_reserve_kbytes affects this policy. 664 665 This feature can be very useful because there are a lot of 666 programs that malloc() huge amounts of memory "just-in-case" 667 and don't use much of it. 668 669 The default value is 0. 670 671 See Documentation/vm/overcommit-accounting and 672 mm/mmap.c::__vm_enough_memory() for more information. 673 674 ============================================================== 675 676 overcommit_ratio: 677 678 When overcommit_memory is set to 2, the committed address 679 space is not permitted to exceed swap plus this percentage 680 of physical RAM. See above. 681 682 ============================================================== 683 684 page-cluster 685 686 page-cluster controls the number of pages up to which consecutive pages 687 are read in from swap in a single attempt. This is the swap counterpart 688 to page cache readahead. 689 The mentioned consecutivity is not in terms of virtual/physical addresses, 690 but consecutive on swap space - that means they were swapped out together. 691 692 It is a logarithmic value - setting it to zero means "1 page", setting 693 it to 1 means "2 pages", setting it to 2 means "4 pages", etc. 694 Zero disables swap readahead completely. 695 696 The default value is three (eight pages at a time). There may be some 697 small benefits in tuning this to a different value if your workload is 698 swap-intensive. 699 700 Lower values mean lower latencies for initial faults, but at the same time 701 extra faults and I/O delays for following faults if they would have been part of 702 that consecutive pages readahead would have brought in. 703 704 ============================================================= 705 706 panic_on_oom 707 708 This enables or disables panic on out-of-memory feature. 709 710 If this is set to 0, the kernel will kill some rogue process, 711 called oom_killer. Usually, oom_killer can kill rogue processes and 712 system will survive. 713 714 If this is set to 1, the kernel panics when out-of-memory happens. 715 However, if a process limits using nodes by mempolicy/cpusets, 716 and those nodes become memory exhaustion status, one process 717 may be killed by oom-killer. No panic occurs in this case. 718 Because other nodes' memory may be free. This means system total status 719 may be not fatal yet. 720 721 If this is set to 2, the kernel panics compulsorily even on the 722 above-mentioned. Even oom happens under memory cgroup, the whole 723 system panics. 724 725 The default value is 0. 726 1 and 2 are for failover of clustering. Please select either 727 according to your policy of failover. 728 panic_on_oom=2+kdump gives you very strong tool to investigate 729 why oom happens. You can get snapshot. 730 731 ============================================================= 732 733 percpu_pagelist_fraction 734 735 This is the fraction of pages at most (high mark pcp->high) in each zone that 736 are allocated for each per cpu page list. The min value for this is 8. It 737 means that we don't allow more than 1/8th of pages in each zone to be 738 allocated in any single per_cpu_pagelist. This entry only changes the value 739 of hot per cpu pagelists. User can specify a number like 100 to allocate 740 1/100th of each zone to each per cpu page list. 741 742 The batch value of each per cpu pagelist is also updated as a result. It is 743 set to pcp->high/4. The upper limit of batch is (PAGE_SHIFT * 8) 744 745 The initial value is zero. Kernel does not use this value at boot time to set 746 the high water marks for each per cpu page list. If the user writes '0' to this 747 sysctl, it will revert to this default behavior. 748 749 ============================================================== 750 751 stat_interval 752 753 The time interval between which vm statistics are updated. The default 754 is 1 second. 755 756 ============================================================== 757 758 stat_refresh 759 760 Any read or write (by root only) flushes all the per-cpu vm statistics 761 into their global totals, for more accurate reports when testing 762 e.g. cat /proc/sys/vm/stat_refresh /proc/meminfo 763 764 As a side-effect, it also checks for negative totals (elsewhere reported 765 as 0) and "fails" with EINVAL if any are found, with a warning in dmesg. 766 (At time of writing, a few stats are known sometimes to be found negative, 767 with no ill effects: errors and warnings on these stats are suppressed.) 768 769 ============================================================== 770 771 numa_stat 772 773 This interface allows runtime configuration of numa statistics. 774 775 When page allocation performance becomes a bottleneck and you can tolerate 776 some possible tool breakage and decreased numa counter precision, you can 777 do: 778 echo 0 > /proc/sys/vm/numa_stat 779 780 When page allocation performance is not a bottleneck and you want all 781 tooling to work, you can do: 782 echo 1 > /proc/sys/vm/numa_stat 783 784 ============================================================== 785 786 swappiness 787 788 This control is used to define how aggressive the kernel will swap 789 memory pages. Higher values will increase aggressiveness, lower values 790 decrease the amount of swap. A value of 0 instructs the kernel not to 791 initiate swap until the amount of free and file-backed pages is less 792 than the high water mark in a zone. 793 794 The default value is 60. 795 796 ============================================================== 797 798 - user_reserve_kbytes 799 800 When overcommit_memory is set to 2, "never overcommit" mode, reserve 801 min(3% of current process size, user_reserve_kbytes) of free memory. 802 This is intended to prevent a user from starting a single memory hogging 803 process, such that they cannot recover (kill the hog). 804 805 user_reserve_kbytes defaults to min(3% of the current process size, 128MB). 806 807 If this is reduced to zero, then the user will be allowed to allocate 808 all free memory with a single process, minus admin_reserve_kbytes. 809 Any subsequent attempts to execute a command will result in 810 "fork: Cannot allocate memory". 811 812 Changing this takes effect whenever an application requests memory. 813 814 ============================================================== 815 816 vfs_cache_pressure 817 ------------------ 818 819 This percentage value controls the tendency of the kernel to reclaim 820 the memory which is used for caching of directory and inode objects. 821 822 At the default value of vfs_cache_pressure=100 the kernel will attempt to 823 reclaim dentries and inodes at a "fair" rate with respect to pagecache and 824 swapcache reclaim. Decreasing vfs_cache_pressure causes the kernel to prefer 825 to retain dentry and inode caches. When vfs_cache_pressure=0, the kernel will 826 never reclaim dentries and inodes due to memory pressure and this can easily 827 lead to out-of-memory conditions. Increasing vfs_cache_pressure beyond 100 828 causes the kernel to prefer to reclaim dentries and inodes. 829 830 Increasing vfs_cache_pressure significantly beyond 100 may have negative 831 performance impact. Reclaim code needs to take various locks to find freeable 832 directory and inode objects. With vfs_cache_pressure=1000, it will look for 833 ten times more freeable objects than there are. 834 835 ============================================================= 836 837 watermark_scale_factor: 838 839 This factor controls the aggressiveness of kswapd. It defines the 840 amount of memory left in a node/system before kswapd is woken up and 841 how much memory needs to be free before kswapd goes back to sleep. 842 843 The unit is in fractions of 10,000. The default value of 10 means the 844 distances between watermarks are 0.1% of the available memory in the 845 node/system. The maximum value is 1000, or 10% of memory. 846 847 A high rate of threads entering direct reclaim (allocstall) or kswapd 848 going to sleep prematurely (kswapd_low_wmark_hit_quickly) can indicate 849 that the number of free pages kswapd maintains for latency reasons is 850 too small for the allocation bursts occurring in the system. This knob 851 can then be used to tune kswapd aggressiveness accordingly. 852 853 ============================================================== 854 855 zone_reclaim_mode: 856 857 Zone_reclaim_mode allows someone to set more or less aggressive approaches to 858 reclaim memory when a zone runs out of memory. If it is set to zero then no 859 zone reclaim occurs. Allocations will be satisfied from other zones / nodes 860 in the system. 861 862 This is value ORed together of 863 864 1 = Zone reclaim on 865 2 = Zone reclaim writes dirty pages out 866 4 = Zone reclaim swaps pages 867 868 zone_reclaim_mode is disabled by default. For file servers or workloads 869 that benefit from having their data cached, zone_reclaim_mode should be 870 left disabled as the caching effect is likely to be more important than 871 data locality. 872 873 zone_reclaim may be enabled if it's known that the workload is partitioned 874 such that each partition fits within a NUMA node and that accessing remote 875 memory would cause a measurable performance reduction. The page allocator 876 will then reclaim easily reusable pages (those page cache pages that are 877 currently not used) before allocating off node pages. 878 879 Allowing zone reclaim to write out pages stops processes that are 880 writing large amounts of data from dirtying pages on other nodes. Zone 881 reclaim will write out dirty pages if a zone fills up and so effectively 882 throttle the process. This may decrease the performance of a single process 883 since it cannot use all of system memory to buffer the outgoing writes 884 anymore but it preserve the memory on other nodes so that the performance 885 of other processes running on other nodes will not be affected. 886 887 Allowing regular swap effectively restricts allocations to the local 888 node unless explicitly overridden by memory policies or cpuset 889 configurations. 890 891 ============ End of Document =================================