1 =============================== 1 ===============================
2 Documentation for /proc/sys/vm/ 2 Documentation for /proc/sys/vm/
3 =============================== 3 ===============================
4 4
5 kernel version 2.6.29 5 kernel version 2.6.29
6 6
7 Copyright (c) 1998, 1999, Rik van Riel <riel@n 7 Copyright (c) 1998, 1999, Rik van Riel <riel@nl.linux.org>
8 8
9 Copyright (c) 2008 Peter W. Morreale <p 9 Copyright (c) 2008 Peter W. Morreale <pmorreale@novell.com>
10 10
11 For general info and legal blurb, please look 11 For general info and legal blurb, please look in index.rst.
12 12
13 ---------------------------------------------- 13 ------------------------------------------------------------------------------
14 14
15 This file contains the documentation for the s 15 This file contains the documentation for the sysctl files in
16 /proc/sys/vm and is valid for Linux kernel ver 16 /proc/sys/vm and is valid for Linux kernel version 2.6.29.
17 17
18 The files in this directory can be used to tun 18 The files in this directory can be used to tune the operation
19 of the virtual memory (VM) subsystem of the Li 19 of the virtual memory (VM) subsystem of the Linux kernel and
20 the writeout of dirty data to disk. 20 the writeout of dirty data to disk.
21 21
22 Default values and initialization routines for 22 Default values and initialization routines for most of these
23 files can be found in mm/swap.c. 23 files can be found in mm/swap.c.
24 24
25 Currently, these files are in /proc/sys/vm: 25 Currently, these files are in /proc/sys/vm:
26 26
27 - admin_reserve_kbytes 27 - admin_reserve_kbytes
28 - compact_memory 28 - compact_memory
29 - compaction_proactiveness 29 - compaction_proactiveness
30 - compact_unevictable_allowed 30 - compact_unevictable_allowed
31 - dirty_background_bytes 31 - dirty_background_bytes
32 - dirty_background_ratio 32 - dirty_background_ratio
33 - dirty_bytes 33 - dirty_bytes
34 - dirty_expire_centisecs 34 - dirty_expire_centisecs
35 - dirty_ratio 35 - dirty_ratio
36 - dirtytime_expire_seconds 36 - dirtytime_expire_seconds
37 - dirty_writeback_centisecs 37 - dirty_writeback_centisecs
38 - drop_caches 38 - drop_caches
39 - enable_soft_offline <<
40 - extfrag_threshold 39 - extfrag_threshold
41 - highmem_is_dirtyable 40 - highmem_is_dirtyable
42 - hugetlb_shm_group 41 - hugetlb_shm_group
43 - laptop_mode 42 - laptop_mode
44 - legacy_va_layout 43 - legacy_va_layout
45 - lowmem_reserve_ratio 44 - lowmem_reserve_ratio
46 - max_map_count 45 - max_map_count
47 - mem_profiling (only if CONFIG_MEM_AL <<
48 - memory_failure_early_kill 46 - memory_failure_early_kill
49 - memory_failure_recovery 47 - memory_failure_recovery
50 - min_free_kbytes 48 - min_free_kbytes
51 - min_slab_ratio 49 - min_slab_ratio
52 - min_unmapped_ratio 50 - min_unmapped_ratio
53 - mmap_min_addr 51 - mmap_min_addr
54 - mmap_rnd_bits 52 - mmap_rnd_bits
55 - mmap_rnd_compat_bits 53 - mmap_rnd_compat_bits
56 - nr_hugepages 54 - nr_hugepages
57 - nr_hugepages_mempolicy 55 - nr_hugepages_mempolicy
58 - nr_overcommit_hugepages 56 - nr_overcommit_hugepages
59 - nr_trim_pages (only if CONFIG_MMU=n) 57 - nr_trim_pages (only if CONFIG_MMU=n)
60 - numa_zonelist_order 58 - numa_zonelist_order
61 - oom_dump_tasks 59 - oom_dump_tasks
62 - oom_kill_allocating_task 60 - oom_kill_allocating_task
63 - overcommit_kbytes 61 - overcommit_kbytes
64 - overcommit_memory 62 - overcommit_memory
65 - overcommit_ratio 63 - overcommit_ratio
66 - page-cluster 64 - page-cluster
67 - page_lock_unfairness <<
68 - panic_on_oom 65 - panic_on_oom
69 - percpu_pagelist_high_fraction 66 - percpu_pagelist_high_fraction
70 - stat_interval 67 - stat_interval
71 - stat_refresh 68 - stat_refresh
72 - numa_stat 69 - numa_stat
73 - swappiness 70 - swappiness
74 - unprivileged_userfaultfd 71 - unprivileged_userfaultfd
75 - user_reserve_kbytes 72 - user_reserve_kbytes
76 - vfs_cache_pressure 73 - vfs_cache_pressure
77 - watermark_boost_factor 74 - watermark_boost_factor
78 - watermark_scale_factor 75 - watermark_scale_factor
79 - zone_reclaim_mode 76 - zone_reclaim_mode
80 77
81 78
82 admin_reserve_kbytes 79 admin_reserve_kbytes
83 ==================== 80 ====================
84 81
85 The amount of free memory in the system that s 82 The amount of free memory in the system that should be reserved for users
86 with the capability cap_sys_admin. 83 with the capability cap_sys_admin.
87 84
88 admin_reserve_kbytes defaults to min(3% of fre 85 admin_reserve_kbytes defaults to min(3% of free pages, 8MB)
89 86
90 That should provide enough for the admin to lo 87 That should provide enough for the admin to log in and kill a process,
91 if necessary, under the default overcommit 'gu 88 if necessary, under the default overcommit 'guess' mode.
92 89
93 Systems running under overcommit 'never' shoul 90 Systems running under overcommit 'never' should increase this to account
94 for the full Virtual Memory Size of programs u 91 for the full Virtual Memory Size of programs used to recover. Otherwise,
95 root may not be able to log in to recover the 92 root may not be able to log in to recover the system.
96 93
97 How do you calculate a minimum useful reserve? 94 How do you calculate a minimum useful reserve?
98 95
99 sshd or login + bash (or some other shell) + t 96 sshd or login + bash (or some other shell) + top (or ps, kill, etc.)
100 97
101 For overcommit 'guess', we can sum resident se 98 For overcommit 'guess', we can sum resident set sizes (RSS).
102 On x86_64 this is about 8MB. 99 On x86_64 this is about 8MB.
103 100
104 For overcommit 'never', we can take the max of 101 For overcommit 'never', we can take the max of their virtual sizes (VSZ)
105 and add the sum of their RSS. 102 and add the sum of their RSS.
106 On x86_64 this is about 128MB. 103 On x86_64 this is about 128MB.
107 104
108 Changing this takes effect whenever an applica 105 Changing this takes effect whenever an application requests memory.
109 106
110 107
111 compact_memory 108 compact_memory
112 ============== 109 ==============
113 110
114 Available only when CONFIG_COMPACTION is set. 111 Available only when CONFIG_COMPACTION is set. When 1 is written to the file,
115 all zones are compacted such that free memory 112 all zones are compacted such that free memory is available in contiguous
116 blocks where possible. This can be important f 113 blocks where possible. This can be important for example in the allocation of
117 huge pages although processes will also direct 114 huge pages although processes will also directly compact memory as required.
118 115
119 compaction_proactiveness 116 compaction_proactiveness
120 ======================== 117 ========================
121 118
122 This tunable takes a value in the range [0, 10 119 This tunable takes a value in the range [0, 100] with a default value of
123 20. This tunable determines how aggressively c 120 20. This tunable determines how aggressively compaction is done in the
124 background. Write of a non zero value to this 121 background. Write of a non zero value to this tunable will immediately
125 trigger the proactive compaction. Setting it t 122 trigger the proactive compaction. Setting it to 0 disables proactive compaction.
126 123
127 Note that compaction has a non-trivial system- 124 Note that compaction has a non-trivial system-wide impact as pages
128 belonging to different processes are moved aro 125 belonging to different processes are moved around, which could also lead
129 to latency spikes in unsuspecting applications 126 to latency spikes in unsuspecting applications. The kernel employs
130 various heuristics to avoid wasting CPU cycles 127 various heuristics to avoid wasting CPU cycles if it detects that
131 proactive compaction is not being effective. 128 proactive compaction is not being effective.
132 129
133 Be careful when setting it to extreme values l 130 Be careful when setting it to extreme values like 100, as that may
134 cause excessive background compaction activity 131 cause excessive background compaction activity.
135 132
136 compact_unevictable_allowed 133 compact_unevictable_allowed
137 =========================== 134 ===========================
138 135
139 Available only when CONFIG_COMPACTION is set. 136 Available only when CONFIG_COMPACTION is set. When set to 1, compaction is
140 allowed to examine the unevictable lru (mlocke 137 allowed to examine the unevictable lru (mlocked pages) for pages to compact.
141 This should be used on systems where stalls fo 138 This should be used on systems where stalls for minor page faults are an
142 acceptable trade for large contiguous free mem 139 acceptable trade for large contiguous free memory. Set to 0 to prevent
143 compaction from moving pages that are unevicta 140 compaction from moving pages that are unevictable. Default value is 1.
144 On CONFIG_PREEMPT_RT the default value is 0 in 141 On CONFIG_PREEMPT_RT the default value is 0 in order to avoid a page fault, due
145 to compaction, which would block the task from 142 to compaction, which would block the task from becoming active until the fault
146 is resolved. 143 is resolved.
147 144
148 145
149 dirty_background_bytes 146 dirty_background_bytes
150 ====================== 147 ======================
151 148
152 Contains the amount of dirty memory at which t 149 Contains the amount of dirty memory at which the background kernel
153 flusher threads will start writeback. 150 flusher threads will start writeback.
154 151
155 Note: 152 Note:
156 dirty_background_bytes is the counterpart of 153 dirty_background_bytes is the counterpart of dirty_background_ratio. Only
157 one of them may be specified at a time. When 154 one of them may be specified at a time. When one sysctl is written it is
158 immediately taken into account to evaluate t 155 immediately taken into account to evaluate the dirty memory limits and the
159 other appears as 0 when read. 156 other appears as 0 when read.
160 157
161 158
162 dirty_background_ratio 159 dirty_background_ratio
163 ====================== 160 ======================
164 161
165 Contains, as a percentage of total available m 162 Contains, as a percentage of total available memory that contains free pages
166 and reclaimable pages, the number of pages at 163 and reclaimable pages, the number of pages at which the background kernel
167 flusher threads will start writing out dirty d 164 flusher threads will start writing out dirty data.
168 165
169 The total available memory is not equal to tot 166 The total available memory is not equal to total system memory.
170 167
171 168
172 dirty_bytes 169 dirty_bytes
173 =========== 170 ===========
174 171
175 Contains the amount of dirty memory at which a 172 Contains the amount of dirty memory at which a process generating disk writes
176 will itself start writeback. 173 will itself start writeback.
177 174
178 Note: dirty_bytes is the counterpart of dirty_ 175 Note: dirty_bytes is the counterpart of dirty_ratio. Only one of them may be
179 specified at a time. When one sysctl is writte 176 specified at a time. When one sysctl is written it is immediately taken into
180 account to evaluate the dirty memory limits an 177 account to evaluate the dirty memory limits and the other appears as 0 when
181 read. 178 read.
182 179
183 Note: the minimum value allowed for dirty_byte 180 Note: the minimum value allowed for dirty_bytes is two pages (in bytes); any
184 value lower than this limit will be ignored an 181 value lower than this limit will be ignored and the old configuration will be
185 retained. 182 retained.
186 183
187 184
188 dirty_expire_centisecs 185 dirty_expire_centisecs
189 ====================== 186 ======================
190 187
191 This tunable is used to define when dirty data 188 This tunable is used to define when dirty data is old enough to be eligible
192 for writeout by the kernel flusher threads. I 189 for writeout by the kernel flusher threads. It is expressed in 100'ths
193 of a second. Data which has been dirty in-mem 190 of a second. Data which has been dirty in-memory for longer than this
194 interval will be written out next time a flush 191 interval will be written out next time a flusher thread wakes up.
195 192
196 193
197 dirty_ratio 194 dirty_ratio
198 =========== 195 ===========
199 196
200 Contains, as a percentage of total available m 197 Contains, as a percentage of total available memory that contains free pages
201 and reclaimable pages, the number of pages at 198 and reclaimable pages, the number of pages at which a process which is
202 generating disk writes will itself start writi 199 generating disk writes will itself start writing out dirty data.
203 200
204 The total available memory is not equal to tot 201 The total available memory is not equal to total system memory.
205 202
206 203
207 dirtytime_expire_seconds 204 dirtytime_expire_seconds
208 ======================== 205 ========================
209 206
210 When a lazytime inode is constantly having its 207 When a lazytime inode is constantly having its pages dirtied, the inode with
211 an updated timestamp will never get chance to 208 an updated timestamp will never get chance to be written out. And, if the
212 only thing that has happened on the file syste 209 only thing that has happened on the file system is a dirtytime inode caused
213 by an atime update, a worker will be scheduled 210 by an atime update, a worker will be scheduled to make sure that inode
214 eventually gets pushed out to disk. This tuna 211 eventually gets pushed out to disk. This tunable is used to define when dirty
215 inode is old enough to be eligible for writeba 212 inode is old enough to be eligible for writeback by the kernel flusher threads.
216 And, it is also used as the interval to wakeup 213 And, it is also used as the interval to wakeup dirtytime_writeback thread.
217 214
218 215
219 dirty_writeback_centisecs 216 dirty_writeback_centisecs
220 ========================= 217 =========================
221 218
222 The kernel flusher threads will periodically w 219 The kernel flusher threads will periodically wake up and write `old` data
223 out to disk. This tunable expresses the inter 220 out to disk. This tunable expresses the interval between those wakeups, in
224 100'ths of a second. 221 100'ths of a second.
225 222
226 Setting this to zero disables periodic writeba 223 Setting this to zero disables periodic writeback altogether.
227 224
228 225
229 drop_caches 226 drop_caches
230 =========== 227 ===========
231 228
232 Writing to this will cause the kernel to drop 229 Writing to this will cause the kernel to drop clean caches, as well as
233 reclaimable slab objects like dentries and ino 230 reclaimable slab objects like dentries and inodes. Once dropped, their
234 memory becomes free. 231 memory becomes free.
235 232
236 To free pagecache:: 233 To free pagecache::
237 234
238 echo 1 > /proc/sys/vm/drop_caches 235 echo 1 > /proc/sys/vm/drop_caches
239 236
240 To free reclaimable slab objects (includes den 237 To free reclaimable slab objects (includes dentries and inodes)::
241 238
242 echo 2 > /proc/sys/vm/drop_caches 239 echo 2 > /proc/sys/vm/drop_caches
243 240
244 To free slab objects and pagecache:: 241 To free slab objects and pagecache::
245 242
246 echo 3 > /proc/sys/vm/drop_caches 243 echo 3 > /proc/sys/vm/drop_caches
247 244
248 This is a non-destructive operation and will n 245 This is a non-destructive operation and will not free any dirty objects.
249 To increase the number of objects freed by thi 246 To increase the number of objects freed by this operation, the user may run
250 `sync` prior to writing to /proc/sys/vm/drop_c 247 `sync` prior to writing to /proc/sys/vm/drop_caches. This will minimize the
251 number of dirty objects on the system and crea 248 number of dirty objects on the system and create more candidates to be
252 dropped. 249 dropped.
253 250
254 This file is not a means to control the growth 251 This file is not a means to control the growth of the various kernel caches
255 (inodes, dentries, pagecache, etc...) These o 252 (inodes, dentries, pagecache, etc...) These objects are automatically
256 reclaimed by the kernel when memory is needed 253 reclaimed by the kernel when memory is needed elsewhere on the system.
257 254
258 Use of this file can cause performance problem 255 Use of this file can cause performance problems. Since it discards cached
259 objects, it may cost a significant amount of I 256 objects, it may cost a significant amount of I/O and CPU to recreate the
260 dropped objects, especially if they were under 257 dropped objects, especially if they were under heavy use. Because of this,
261 use outside of a testing or debugging environm 258 use outside of a testing or debugging environment is not recommended.
262 259
263 You may see informational messages in your ker 260 You may see informational messages in your kernel log when this file is
264 used:: 261 used::
265 262
266 cat (1234): drop_caches: 3 263 cat (1234): drop_caches: 3
267 264
268 These are informational only. They do not mea 265 These are informational only. They do not mean that anything is wrong
269 with your system. To disable them, echo 4 (bi 266 with your system. To disable them, echo 4 (bit 2) into drop_caches.
270 267
271 enable_soft_offline <<
272 =================== <<
273 Correctable memory errors are very common on s <<
274 solution for memory pages having (excessive) c <<
275 <<
276 For different types of page, soft-offline has <<
277 <<
278 - For a raw error page, soft-offline migrates <<
279 a new raw page. <<
280 <<
281 - For a page that is part of a transparent hug <<
282 transparent hugepage into raw pages, then mi <<
283 As a result, user is transparently backed by <<
284 memory access performance. <<
285 <<
286 - For a page that is part of a HugeTLB hugepag <<
287 the entire HugeTLB hugepage, during which a <<
288 as migration target. Then the original huge <<
289 pages without compensation, reducing the cap <<
290 <<
291 It is user's call to choose between reliabilit <<
292 physical memory) vs performance / capacity imp <<
293 HugeTLB cases. <<
294 <<
295 For all architectures, enable_soft_offline con <<
296 memory pages. When set to 1, kernel attempts <<
297 whenever it thinks needed. When set to 0, ker <<
298 the request to soft offline the pages. Its de <<
299 <<
300 It is worth mentioning that after setting enab <<
301 following requests to soft offline pages will <<
302 <<
303 - Request to soft offline pages from RAS Corre <<
304 <<
305 - On ARM, the request to soft offline pages fr <<
306 <<
307 - On PARISC, the request to soft offline pages <<
308 268
309 extfrag_threshold 269 extfrag_threshold
310 ================= 270 =================
311 271
312 This parameter affects whether the kernel will 272 This parameter affects whether the kernel will compact memory or direct
313 reclaim to satisfy a high-order allocation. Th 273 reclaim to satisfy a high-order allocation. The extfrag/extfrag_index file in
314 debugfs shows what the fragmentation index for 274 debugfs shows what the fragmentation index for each order is in each zone in
315 the system. Values tending towards 0 imply all 275 the system. Values tending towards 0 imply allocations would fail due to lack
316 of memory, values towards 1000 imply failures 276 of memory, values towards 1000 imply failures are due to fragmentation and -1
317 implies that the allocation will succeed as lo 277 implies that the allocation will succeed as long as watermarks are met.
318 278
319 The kernel will not compact memory in a zone i 279 The kernel will not compact memory in a zone if the
320 fragmentation index is <= extfrag_threshold. T 280 fragmentation index is <= extfrag_threshold. The default value is 500.
321 281
322 282
323 highmem_is_dirtyable 283 highmem_is_dirtyable
324 ==================== 284 ====================
325 285
326 Available only for systems with CONFIG_HIGHMEM 286 Available only for systems with CONFIG_HIGHMEM enabled (32b systems).
327 287
328 This parameter controls whether the high memor 288 This parameter controls whether the high memory is considered for dirty
329 writers throttling. This is not the case by d 289 writers throttling. This is not the case by default which means that
330 only the amount of memory directly visible/usa 290 only the amount of memory directly visible/usable by the kernel can
331 be dirtied. As a result, on systems with a lar 291 be dirtied. As a result, on systems with a large amount of memory and
332 lowmem basically depleted writers might be thr 292 lowmem basically depleted writers might be throttled too early and
333 streaming writes can get very slow. 293 streaming writes can get very slow.
334 294
335 Changing the value to non zero would allow mor 295 Changing the value to non zero would allow more memory to be dirtied
336 and thus allow writers to write more data whic 296 and thus allow writers to write more data which can be flushed to the
337 storage more effectively. Note this also comes 297 storage more effectively. Note this also comes with a risk of pre-mature
338 OOM killer because some writers (e.g. direct b 298 OOM killer because some writers (e.g. direct block device writes) can
339 only use the low memory and they can fill it u 299 only use the low memory and they can fill it up with dirty data without
340 any throttling. 300 any throttling.
341 301
342 302
343 hugetlb_shm_group 303 hugetlb_shm_group
344 ================= 304 =================
345 305
346 hugetlb_shm_group contains group id that is al 306 hugetlb_shm_group contains group id that is allowed to create SysV
347 shared memory segment using hugetlb page. 307 shared memory segment using hugetlb page.
348 308
349 309
350 laptop_mode 310 laptop_mode
351 =========== 311 ===========
352 312
353 laptop_mode is a knob that controls "laptop mo 313 laptop_mode is a knob that controls "laptop mode". All the things that are
354 controlled by this knob are discussed in Docum 314 controlled by this knob are discussed in Documentation/admin-guide/laptops/laptop-mode.rst.
355 315
356 316
357 legacy_va_layout 317 legacy_va_layout
358 ================ 318 ================
359 319
360 If non-zero, this sysctl disables the new 32-b 320 If non-zero, this sysctl disables the new 32-bit mmap layout - the kernel
361 will use the legacy (2.4) layout for all proce 321 will use the legacy (2.4) layout for all processes.
362 322
363 323
364 lowmem_reserve_ratio 324 lowmem_reserve_ratio
365 ==================== 325 ====================
366 326
367 For some specialised workloads on highmem mach 327 For some specialised workloads on highmem machines it is dangerous for
368 the kernel to allow process memory to be alloc 328 the kernel to allow process memory to be allocated from the "lowmem"
369 zone. This is because that memory could then 329 zone. This is because that memory could then be pinned via the mlock()
370 system call, or by unavailability of swapspace 330 system call, or by unavailability of swapspace.
371 331
372 And on large highmem machines this lack of rec 332 And on large highmem machines this lack of reclaimable lowmem memory
373 can be fatal. 333 can be fatal.
374 334
375 So the Linux page allocator has a mechanism wh 335 So the Linux page allocator has a mechanism which prevents allocations
376 which *could* use highmem from using too much 336 which *could* use highmem from using too much lowmem. This means that
377 a certain amount of lowmem is defended from th 337 a certain amount of lowmem is defended from the possibility of being
378 captured into pinned user memory. 338 captured into pinned user memory.
379 339
380 (The same argument applies to the old 16 megab 340 (The same argument applies to the old 16 megabyte ISA DMA region. This
381 mechanism will also defend that region from al 341 mechanism will also defend that region from allocations which could use
382 highmem or lowmem). 342 highmem or lowmem).
383 343
384 The `lowmem_reserve_ratio` tunable determines 344 The `lowmem_reserve_ratio` tunable determines how aggressive the kernel is
385 in defending these lower zones. 345 in defending these lower zones.
386 346
387 If you have a machine which uses highmem or IS 347 If you have a machine which uses highmem or ISA DMA and your
388 applications are using mlock(), or if you are 348 applications are using mlock(), or if you are running with no swap then
389 you probably should change the lowmem_reserve_ 349 you probably should change the lowmem_reserve_ratio setting.
390 350
391 The lowmem_reserve_ratio is an array. You can 351 The lowmem_reserve_ratio is an array. You can see them by reading this file::
392 352
393 % cat /proc/sys/vm/lowmem_reserve_rati 353 % cat /proc/sys/vm/lowmem_reserve_ratio
394 256 256 32 354 256 256 32
395 355
396 But, these values are not used directly. The k 356 But, these values are not used directly. The kernel calculates # of protection
397 pages for each zones from them. These are show 357 pages for each zones from them. These are shown as array of protection pages
398 in /proc/zoneinfo like the following. (This is !! 358 in /proc/zoneinfo like followings. (This is an example of x86-64 box).
399 Each zone has an array of protection pages lik 359 Each zone has an array of protection pages like this::
400 360
401 Node 0, zone DMA 361 Node 0, zone DMA
402 pages free 1355 362 pages free 1355
403 min 3 363 min 3
404 low 3 364 low 3
405 high 4 365 high 4
406 : 366 :
407 : 367 :
408 numa_other 0 368 numa_other 0
409 protection: (0, 2004, 2004, 2004) 369 protection: (0, 2004, 2004, 2004)
410 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 370 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
411 pagesets 371 pagesets
412 cpu: 0 pcp: 0 372 cpu: 0 pcp: 0
413 : 373 :
414 374
415 These protections are added to score to judge 375 These protections are added to score to judge whether this zone should be used
416 for page allocation or should be reclaimed. 376 for page allocation or should be reclaimed.
417 377
418 In this example, if normal pages (index=2) are 378 In this example, if normal pages (index=2) are required to this DMA zone and
419 watermark[WMARK_HIGH] is used for watermark, t 379 watermark[WMARK_HIGH] is used for watermark, the kernel judges this zone should
420 not be used because pages_free(1355) is smalle 380 not be used because pages_free(1355) is smaller than watermark + protection[2]
421 (4 + 2004 = 2008). If this protection value is 381 (4 + 2004 = 2008). If this protection value is 0, this zone would be used for
422 normal page requirement. If requirement is DMA 382 normal page requirement. If requirement is DMA zone(index=0), protection[0]
423 (=0) is used. 383 (=0) is used.
424 384
425 zone[i]'s protection[j] is calculated by follo 385 zone[i]'s protection[j] is calculated by following expression::
426 386
427 (i < j): 387 (i < j):
428 zone[i]->protection[j] 388 zone[i]->protection[j]
429 = (total sums of managed_pages from zone[i 389 = (total sums of managed_pages from zone[i+1] to zone[j] on the node)
430 / lowmem_reserve_ratio[i]; 390 / lowmem_reserve_ratio[i];
431 (i = j): 391 (i = j):
432 (should not be protected. = 0; 392 (should not be protected. = 0;
433 (i > j): 393 (i > j):
434 (not necessary, but looks 0) 394 (not necessary, but looks 0)
435 395
436 The default values of lowmem_reserve_ratio[i] 396 The default values of lowmem_reserve_ratio[i] are
437 397
438 === ==================================== 398 === ====================================
439 256 (if zone[i] means DMA or DMA32 zone) 399 256 (if zone[i] means DMA or DMA32 zone)
440 32 (others) 400 32 (others)
441 === ==================================== 401 === ====================================
442 402
443 As above expression, they are reciprocal numbe 403 As above expression, they are reciprocal number of ratio.
444 256 means 1/256. # of protection pages becomes 404 256 means 1/256. # of protection pages becomes about "0.39%" of total managed
445 pages of higher zones on the node. 405 pages of higher zones on the node.
446 406
447 If you would like to protect more pages, small 407 If you would like to protect more pages, smaller values are effective.
448 The minimum value is 1 (1/1 -> 100%). The valu 408 The minimum value is 1 (1/1 -> 100%). The value less than 1 completely
449 disables protection of the pages. 409 disables protection of the pages.
450 410
451 411
452 max_map_count: 412 max_map_count:
453 ============== 413 ==============
454 414
455 This file contains the maximum number of memor 415 This file contains the maximum number of memory map areas a process
456 may have. Memory map areas are used as a side- 416 may have. Memory map areas are used as a side-effect of calling
457 malloc, directly by mmap, mprotect, and madvis 417 malloc, directly by mmap, mprotect, and madvise, and also when loading
458 shared libraries. 418 shared libraries.
459 419
460 While most applications need less than a thous 420 While most applications need less than a thousand maps, certain
461 programs, particularly malloc debuggers, may c 421 programs, particularly malloc debuggers, may consume lots of them,
462 e.g., up to one or two maps per allocation. 422 e.g., up to one or two maps per allocation.
463 423
464 The default value is 65530. 424 The default value is 65530.
465 425
466 426
467 mem_profiling <<
468 ============== <<
469 <<
470 Enable memory profiling (when CONFIG_MEM_ALLOC <<
471 <<
472 1: Enable memory profiling. <<
473 <<
474 0: Disable memory profiling. <<
475 <<
476 Enabling memory profiling introduces a small p <<
477 memory allocations. <<
478 <<
479 The default value depends on CONFIG_MEM_ALLOC_ <<
480 <<
481 <<
482 memory_failure_early_kill: 427 memory_failure_early_kill:
483 ========================== 428 ==========================
484 429
485 Control how to kill processes when uncorrected 430 Control how to kill processes when uncorrected memory error (typically
486 a 2bit error in a memory module) is detected i 431 a 2bit error in a memory module) is detected in the background by hardware
487 that cannot be handled by the kernel. In some 432 that cannot be handled by the kernel. In some cases (like the page
488 still having a valid copy on disk) the kernel 433 still having a valid copy on disk) the kernel will handle the failure
489 transparently without affecting any applicatio 434 transparently without affecting any applications. But if there is
490 no other up-to-date copy of the data it will k !! 435 no other uptodate copy of the data it will kill to prevent any data
491 corruptions from propagating. 436 corruptions from propagating.
492 437
493 1: Kill all processes that have the corrupted 438 1: Kill all processes that have the corrupted and not reloadable page mapped
494 as soon as the corruption is detected. Note t 439 as soon as the corruption is detected. Note this is not supported
495 for a few types of pages, like kernel internal 440 for a few types of pages, like kernel internally allocated data or
496 the swap cache, but works for the majority of 441 the swap cache, but works for the majority of user pages.
497 442
498 0: Only unmap the corrupted page from all proc 443 0: Only unmap the corrupted page from all processes and only kill a process
499 who tries to access it. 444 who tries to access it.
500 445
501 The kill is done using a catchable SIGBUS with 446 The kill is done using a catchable SIGBUS with BUS_MCEERR_AO, so processes can
502 handle this if they want to. 447 handle this if they want to.
503 448
504 This is only active on architectures/platforms 449 This is only active on architectures/platforms with advanced machine
505 check handling and depends on the hardware cap 450 check handling and depends on the hardware capabilities.
506 451
507 Applications can override this setting individ 452 Applications can override this setting individually with the PR_MCE_KILL prctl
508 453
509 454
510 memory_failure_recovery 455 memory_failure_recovery
511 ======================= 456 =======================
512 457
513 Enable memory failure recovery (when supported 458 Enable memory failure recovery (when supported by the platform)
514 459
515 1: Attempt recovery. 460 1: Attempt recovery.
516 461
517 0: Always panic on a memory failure. 462 0: Always panic on a memory failure.
518 463
519 464
520 min_free_kbytes 465 min_free_kbytes
521 =============== 466 ===============
522 467
523 This is used to force the Linux VM to keep a m 468 This is used to force the Linux VM to keep a minimum number
524 of kilobytes free. The VM uses this number to 469 of kilobytes free. The VM uses this number to compute a
525 watermark[WMARK_MIN] value for each lowmem zon 470 watermark[WMARK_MIN] value for each lowmem zone in the system.
526 Each lowmem zone gets a number of reserved fre 471 Each lowmem zone gets a number of reserved free pages based
527 proportionally on its size. 472 proportionally on its size.
528 473
529 Some minimal amount of memory is needed to sat 474 Some minimal amount of memory is needed to satisfy PF_MEMALLOC
530 allocations; if you set this to lower than 102 475 allocations; if you set this to lower than 1024KB, your system will
531 become subtly broken, and prone to deadlock un 476 become subtly broken, and prone to deadlock under high loads.
532 477
533 Setting this too high will OOM your machine in 478 Setting this too high will OOM your machine instantly.
534 479
535 480
536 min_slab_ratio 481 min_slab_ratio
537 ============== 482 ==============
538 483
539 This is available only on NUMA kernels. 484 This is available only on NUMA kernels.
540 485
541 A percentage of the total pages in each zone. 486 A percentage of the total pages in each zone. On Zone reclaim
542 (fallback from the local zone occurs) slabs wi 487 (fallback from the local zone occurs) slabs will be reclaimed if more
543 than this percentage of pages in a zone are re 488 than this percentage of pages in a zone are reclaimable slab pages.
544 This insures that the slab growth stays under 489 This insures that the slab growth stays under control even in NUMA
545 systems that rarely perform global reclaim. 490 systems that rarely perform global reclaim.
546 491
547 The default is 5 percent. 492 The default is 5 percent.
548 493
549 Note that slab reclaim is triggered in a per z 494 Note that slab reclaim is triggered in a per zone / node fashion.
550 The process of reclaiming slab memory is curre 495 The process of reclaiming slab memory is currently not node specific
551 and may not be fast. 496 and may not be fast.
552 497
553 498
554 min_unmapped_ratio 499 min_unmapped_ratio
555 ================== 500 ==================
556 501
557 This is available only on NUMA kernels. 502 This is available only on NUMA kernels.
558 503
559 This is a percentage of the total pages in eac 504 This is a percentage of the total pages in each zone. Zone reclaim will
560 only occur if more than this percentage of pag 505 only occur if more than this percentage of pages are in a state that
561 zone_reclaim_mode allows to be reclaimed. 506 zone_reclaim_mode allows to be reclaimed.
562 507
563 If zone_reclaim_mode has the value 4 OR'd, the 508 If zone_reclaim_mode has the value 4 OR'd, then the percentage is compared
564 against all file-backed unmapped pages includi 509 against all file-backed unmapped pages including swapcache pages and tmpfs
565 files. Otherwise, only unmapped pages backed b 510 files. Otherwise, only unmapped pages backed by normal files but not tmpfs
566 files and similar are considered. 511 files and similar are considered.
567 512
568 The default is 1 percent. 513 The default is 1 percent.
569 514
570 515
571 mmap_min_addr 516 mmap_min_addr
572 ============= 517 =============
573 518
574 This file indicates the amount of address spac 519 This file indicates the amount of address space which a user process will
575 be restricted from mmapping. Since kernel nul 520 be restricted from mmapping. Since kernel null dereference bugs could
576 accidentally operate based on the information 521 accidentally operate based on the information in the first couple of pages
577 of memory userspace processes should not be al 522 of memory userspace processes should not be allowed to write to them. By
578 default this value is set to 0 and no protecti 523 default this value is set to 0 and no protections will be enforced by the
579 security module. Setting this value to someth 524 security module. Setting this value to something like 64k will allow the
580 vast majority of applications to work correctl 525 vast majority of applications to work correctly and provide defense in depth
581 against future potential kernel bugs. 526 against future potential kernel bugs.
582 527
583 528
584 mmap_rnd_bits 529 mmap_rnd_bits
585 ============= 530 =============
586 531
587 This value can be used to select the number of 532 This value can be used to select the number of bits to use to
588 determine the random offset to the base addres 533 determine the random offset to the base address of vma regions
589 resulting from mmap allocations on architectur 534 resulting from mmap allocations on architectures which support
590 tuning address space randomization. This valu 535 tuning address space randomization. This value will be bounded
591 by the architecture's minimum and maximum supp 536 by the architecture's minimum and maximum supported values.
592 537
593 This value can be changed after boot using the 538 This value can be changed after boot using the
594 /proc/sys/vm/mmap_rnd_bits tunable 539 /proc/sys/vm/mmap_rnd_bits tunable
595 540
596 541
597 mmap_rnd_compat_bits 542 mmap_rnd_compat_bits
598 ==================== 543 ====================
599 544
600 This value can be used to select the number of 545 This value can be used to select the number of bits to use to
601 determine the random offset to the base addres 546 determine the random offset to the base address of vma regions
602 resulting from mmap allocations for applicatio 547 resulting from mmap allocations for applications run in
603 compatibility mode on architectures which supp 548 compatibility mode on architectures which support tuning address
604 space randomization. This value will be bound 549 space randomization. This value will be bounded by the
605 architecture's minimum and maximum supported v 550 architecture's minimum and maximum supported values.
606 551
607 This value can be changed after boot using the 552 This value can be changed after boot using the
608 /proc/sys/vm/mmap_rnd_compat_bits tunable 553 /proc/sys/vm/mmap_rnd_compat_bits tunable
609 554
610 555
611 nr_hugepages 556 nr_hugepages
612 ============ 557 ============
613 558
614 Change the minimum size of the hugepage pool. 559 Change the minimum size of the hugepage pool.
615 560
616 See Documentation/admin-guide/mm/hugetlbpage.r 561 See Documentation/admin-guide/mm/hugetlbpage.rst
617 562
618 563
619 hugetlb_optimize_vmemmap <<
620 ======================== <<
621 <<
622 This knob is not available when the size of 's <<
623 in include/linux/mm_types.h) is not power of t <<
624 result in this). <<
625 <<
626 Enable (set to 1) or disable (set to 0) HugeTL <<
627 <<
628 Once enabled, the vmemmap pages of subsequent <<
629 buddy allocator will be optimized (7 pages per <<
630 per 1GB HugeTLB page), whereas already allocat <<
631 optimized. When those optimized HugeTLB pages <<
632 to the buddy allocator, the vmemmap pages repr <<
633 remapped again and the vmemmap pages discarded <<
634 again. If your use case is that HugeTLB pages <<
635 never explicitly allocating HugeTLB pages with <<
636 'nr_overcommit_hugepages', those overcommitted <<
637 the fly') instead of being pulled from the Hug <<
638 benefits of memory savings against the more ov <<
639 of allocation or freeing HugeTLB pages between <<
640 allocator. Another behavior to note is that i <<
641 pressure, it could prevent the user from freei <<
642 pool to the buddy allocator since the allocati <<
643 failed, you have to retry later if your system <<
644 <<
645 Once disabled, the vmemmap pages of subsequent <<
646 buddy allocator will not be optimized meaning <<
647 time from buddy allocator disappears, whereas <<
648 will not be affected. If you want to make sur <<
649 pages, you can set "nr_hugepages" to 0 first a <<
650 writing 0 to nr_hugepages will make any "in us <<
651 pages. So, those surplus pages are still opti <<
652 in use. You would need to wait for those surp <<
653 there are no optimized pages in the system. <<
654 <<
655 <<
656 nr_hugepages_mempolicy 564 nr_hugepages_mempolicy
657 ====================== 565 ======================
658 566
659 Change the size of the hugepage pool at run-ti 567 Change the size of the hugepage pool at run-time on a specific
660 set of NUMA nodes. 568 set of NUMA nodes.
661 569
662 See Documentation/admin-guide/mm/hugetlbpage.r 570 See Documentation/admin-guide/mm/hugetlbpage.rst
663 571
664 572
665 nr_overcommit_hugepages 573 nr_overcommit_hugepages
666 ======================= 574 =======================
667 575
668 Change the maximum size of the hugepage pool. 576 Change the maximum size of the hugepage pool. The maximum is
669 nr_hugepages + nr_overcommit_hugepages. 577 nr_hugepages + nr_overcommit_hugepages.
670 578
671 See Documentation/admin-guide/mm/hugetlbpage.r 579 See Documentation/admin-guide/mm/hugetlbpage.rst
672 580
673 581
674 nr_trim_pages 582 nr_trim_pages
675 ============= 583 =============
676 584
677 This is available only on NOMMU kernels. 585 This is available only on NOMMU kernels.
678 586
679 This value adjusts the excess page trimming be 587 This value adjusts the excess page trimming behaviour of power-of-2 aligned
680 NOMMU mmap allocations. 588 NOMMU mmap allocations.
681 589
682 A value of 0 disables trimming of allocations 590 A value of 0 disables trimming of allocations entirely, while a value of 1
683 trims excess pages aggressively. Any value >= 591 trims excess pages aggressively. Any value >= 1 acts as the watermark where
684 trimming of allocations is initiated. 592 trimming of allocations is initiated.
685 593
686 The default value is 1. 594 The default value is 1.
687 595
688 See Documentation/admin-guide/mm/nommu-mmap.rs 596 See Documentation/admin-guide/mm/nommu-mmap.rst for more information.
689 597
690 598
691 numa_zonelist_order 599 numa_zonelist_order
692 =================== 600 ===================
693 601
694 This sysctl is only for NUMA and it is depreca 602 This sysctl is only for NUMA and it is deprecated. Anything but
695 Node order will fail! 603 Node order will fail!
696 604
697 'where the memory is allocated from' is contro 605 'where the memory is allocated from' is controlled by zonelists.
698 606
699 (This documentation ignores ZONE_HIGHMEM/ZONE_ 607 (This documentation ignores ZONE_HIGHMEM/ZONE_DMA32 for simple explanation.
700 you may be able to read ZONE_DMA as ZONE_DMA32 608 you may be able to read ZONE_DMA as ZONE_DMA32...)
701 609
702 In non-NUMA case, a zonelist for GFP_KERNEL is 610 In non-NUMA case, a zonelist for GFP_KERNEL is ordered as following.
703 ZONE_NORMAL -> ZONE_DMA 611 ZONE_NORMAL -> ZONE_DMA
704 This means that a memory allocation request fo 612 This means that a memory allocation request for GFP_KERNEL will
705 get memory from ZONE_DMA only when ZONE_NORMAL 613 get memory from ZONE_DMA only when ZONE_NORMAL is not available.
706 614
707 In NUMA case, you can think of following 2 typ 615 In NUMA case, you can think of following 2 types of order.
708 Assume 2 node NUMA and below is zonelist of No 616 Assume 2 node NUMA and below is zonelist of Node(0)'s GFP_KERNEL::
709 617
710 (A) Node(0) ZONE_NORMAL -> Node(0) ZONE_DMA 618 (A) Node(0) ZONE_NORMAL -> Node(0) ZONE_DMA -> Node(1) ZONE_NORMAL
711 (B) Node(0) ZONE_NORMAL -> Node(1) ZONE_NORM 619 (B) Node(0) ZONE_NORMAL -> Node(1) ZONE_NORMAL -> Node(0) ZONE_DMA.
712 620
713 Type(A) offers the best locality for processes 621 Type(A) offers the best locality for processes on Node(0), but ZONE_DMA
714 will be used before ZONE_NORMAL exhaustion. Th 622 will be used before ZONE_NORMAL exhaustion. This increases possibility of
715 out-of-memory(OOM) of ZONE_DMA because ZONE_DM 623 out-of-memory(OOM) of ZONE_DMA because ZONE_DMA is tend to be small.
716 624
717 Type(B) cannot offer the best locality but is 625 Type(B) cannot offer the best locality but is more robust against OOM of
718 the DMA zone. 626 the DMA zone.
719 627
720 Type(A) is called as "Node" order. Type (B) is 628 Type(A) is called as "Node" order. Type (B) is "Zone" order.
721 629
722 "Node order" orders the zonelists by node, the 630 "Node order" orders the zonelists by node, then by zone within each node.
723 Specify "[Nn]ode" for node order 631 Specify "[Nn]ode" for node order
724 632
725 "Zone Order" orders the zonelists by zone type 633 "Zone Order" orders the zonelists by zone type, then by node within each
726 zone. Specify "[Zz]one" for zone order. 634 zone. Specify "[Zz]one" for zone order.
727 635
728 Specify "[Dd]efault" to request automatic conf 636 Specify "[Dd]efault" to request automatic configuration.
729 637
730 On 32-bit, the Normal zone needs to be preserv 638 On 32-bit, the Normal zone needs to be preserved for allocations accessible
731 by the kernel, so "zone" order will be selecte 639 by the kernel, so "zone" order will be selected.
732 640
733 On 64-bit, devices that require DMA32/DMA are 641 On 64-bit, devices that require DMA32/DMA are relatively rare, so "node"
734 order will be selected. 642 order will be selected.
735 643
736 Default order is recommended unless this is ca 644 Default order is recommended unless this is causing problems for your
737 system/application. 645 system/application.
738 646
739 647
740 oom_dump_tasks 648 oom_dump_tasks
741 ============== 649 ==============
742 650
743 Enables a system-wide task dump (excluding ker 651 Enables a system-wide task dump (excluding kernel threads) to be produced
744 when the kernel performs an OOM-killing and in 652 when the kernel performs an OOM-killing and includes such information as
745 pid, uid, tgid, vm size, rss, pgtables_bytes, 653 pid, uid, tgid, vm size, rss, pgtables_bytes, swapents, oom_score_adj
746 score, and name. This is helpful to determine 654 score, and name. This is helpful to determine why the OOM killer was
747 invoked, to identify the rogue task that cause 655 invoked, to identify the rogue task that caused it, and to determine why
748 the OOM killer chose the task it did to kill. 656 the OOM killer chose the task it did to kill.
749 657
750 If this is set to zero, this information is su 658 If this is set to zero, this information is suppressed. On very
751 large systems with thousands of tasks it may n 659 large systems with thousands of tasks it may not be feasible to dump
752 the memory state information for each one. Su 660 the memory state information for each one. Such systems should not
753 be forced to incur a performance penalty in OO 661 be forced to incur a performance penalty in OOM conditions when the
754 information may not be desired. 662 information may not be desired.
755 663
756 If this is set to non-zero, this information i 664 If this is set to non-zero, this information is shown whenever the
757 OOM killer actually kills a memory-hogging tas 665 OOM killer actually kills a memory-hogging task.
758 666
759 The default value is 1 (enabled). 667 The default value is 1 (enabled).
760 668
761 669
762 oom_kill_allocating_task 670 oom_kill_allocating_task
763 ======================== 671 ========================
764 672
765 This enables or disables killing the OOM-trigg 673 This enables or disables killing the OOM-triggering task in
766 out-of-memory situations. 674 out-of-memory situations.
767 675
768 If this is set to zero, the OOM killer will sc 676 If this is set to zero, the OOM killer will scan through the entire
769 tasklist and select a task based on heuristics 677 tasklist and select a task based on heuristics to kill. This normally
770 selects a rogue memory-hogging task that frees 678 selects a rogue memory-hogging task that frees up a large amount of
771 memory when killed. 679 memory when killed.
772 680
773 If this is set to non-zero, the OOM killer sim 681 If this is set to non-zero, the OOM killer simply kills the task that
774 triggered the out-of-memory condition. This a 682 triggered the out-of-memory condition. This avoids the expensive
775 tasklist scan. 683 tasklist scan.
776 684
777 If panic_on_oom is selected, it takes preceden 685 If panic_on_oom is selected, it takes precedence over whatever value
778 is used in oom_kill_allocating_task. 686 is used in oom_kill_allocating_task.
779 687
780 The default value is 0. 688 The default value is 0.
781 689
782 690
783 overcommit_kbytes 691 overcommit_kbytes
784 ================= 692 =================
785 693
786 When overcommit_memory is set to 2, the commit 694 When overcommit_memory is set to 2, the committed address space is not
787 permitted to exceed swap plus this amount of p 695 permitted to exceed swap plus this amount of physical RAM. See below.
788 696
789 Note: overcommit_kbytes is the counterpart of 697 Note: overcommit_kbytes is the counterpart of overcommit_ratio. Only one
790 of them may be specified at a time. Setting on 698 of them may be specified at a time. Setting one disables the other (which
791 then appears as 0 when read). 699 then appears as 0 when read).
792 700
793 701
794 overcommit_memory 702 overcommit_memory
795 ================= 703 =================
796 704
797 This value contains a flag that enables memory 705 This value contains a flag that enables memory overcommitment.
798 706
799 When this flag is 0, the kernel compares the u !! 707 When this flag is 0, the kernel attempts to estimate the amount
800 size against total memory plus swap and reject !! 708 of free memory left when userspace requests more memory.
801 709
802 When this flag is 1, the kernel pretends there 710 When this flag is 1, the kernel pretends there is always enough
803 memory until it actually runs out. 711 memory until it actually runs out.
804 712
805 When this flag is 2, the kernel uses a "never 713 When this flag is 2, the kernel uses a "never overcommit"
806 policy that attempts to prevent any overcommit 714 policy that attempts to prevent any overcommit of memory.
807 Note that user_reserve_kbytes affects this pol 715 Note that user_reserve_kbytes affects this policy.
808 716
809 This feature can be very useful because there 717 This feature can be very useful because there are a lot of
810 programs that malloc() huge amounts of memory 718 programs that malloc() huge amounts of memory "just-in-case"
811 and don't use much of it. 719 and don't use much of it.
812 720
813 The default value is 0. 721 The default value is 0.
814 722
815 See Documentation/mm/overcommit-accounting.rst !! 723 See Documentation/vm/overcommit-accounting.rst and
816 mm/util.c::__vm_enough_memory() for more infor 724 mm/util.c::__vm_enough_memory() for more information.
817 725
818 726
819 overcommit_ratio 727 overcommit_ratio
820 ================ 728 ================
821 729
822 When overcommit_memory is set to 2, the commit 730 When overcommit_memory is set to 2, the committed address
823 space is not permitted to exceed swap plus thi 731 space is not permitted to exceed swap plus this percentage
824 of physical RAM. See above. 732 of physical RAM. See above.
825 733
826 734
827 page-cluster 735 page-cluster
828 ============ 736 ============
829 737
830 page-cluster controls the number of pages up t 738 page-cluster controls the number of pages up to which consecutive pages
831 are read in from swap in a single attempt. Thi 739 are read in from swap in a single attempt. This is the swap counterpart
832 to page cache readahead. 740 to page cache readahead.
833 The mentioned consecutivity is not in terms of 741 The mentioned consecutivity is not in terms of virtual/physical addresses,
834 but consecutive on swap space - that means the 742 but consecutive on swap space - that means they were swapped out together.
835 743
836 It is a logarithmic value - setting it to zero 744 It is a logarithmic value - setting it to zero means "1 page", setting
837 it to 1 means "2 pages", setting it to 2 means 745 it to 1 means "2 pages", setting it to 2 means "4 pages", etc.
838 Zero disables swap readahead completely. 746 Zero disables swap readahead completely.
839 747
840 The default value is three (eight pages at a t 748 The default value is three (eight pages at a time). There may be some
841 small benefits in tuning this to a different v 749 small benefits in tuning this to a different value if your workload is
842 swap-intensive. 750 swap-intensive.
843 751
844 Lower values mean lower latencies for initial 752 Lower values mean lower latencies for initial faults, but at the same time
845 extra faults and I/O delays for following faul 753 extra faults and I/O delays for following faults if they would have been part of
846 that consecutive pages readahead would have br 754 that consecutive pages readahead would have brought in.
847 755
848 756
849 page_lock_unfairness <<
850 ==================== <<
851 <<
852 This value determines the number of times that <<
853 stolen from under a waiter. After the lock is <<
854 specified in this file (default is 5), the "fa <<
855 will apply, and the waiter will only be awaken <<
856 <<
857 panic_on_oom 757 panic_on_oom
858 ============ 758 ============
859 759
860 This enables or disables panic on out-of-memor 760 This enables or disables panic on out-of-memory feature.
861 761
862 If this is set to 0, the kernel will kill some 762 If this is set to 0, the kernel will kill some rogue process,
863 called oom_killer. Usually, oom_killer can ki 763 called oom_killer. Usually, oom_killer can kill rogue processes and
864 system will survive. 764 system will survive.
865 765
866 If this is set to 1, the kernel panics when ou 766 If this is set to 1, the kernel panics when out-of-memory happens.
867 However, if a process limits using nodes by me 767 However, if a process limits using nodes by mempolicy/cpusets,
868 and those nodes become memory exhaustion statu 768 and those nodes become memory exhaustion status, one process
869 may be killed by oom-killer. No panic occurs i 769 may be killed by oom-killer. No panic occurs in this case.
870 Because other nodes' memory may be free. This 770 Because other nodes' memory may be free. This means system total status
871 may be not fatal yet. 771 may be not fatal yet.
872 772
873 If this is set to 2, the kernel panics compuls 773 If this is set to 2, the kernel panics compulsorily even on the
874 above-mentioned. Even oom happens under memory 774 above-mentioned. Even oom happens under memory cgroup, the whole
875 system panics. 775 system panics.
876 776
877 The default value is 0. 777 The default value is 0.
878 778
879 1 and 2 are for failover of clustering. Please 779 1 and 2 are for failover of clustering. Please select either
880 according to your policy of failover. 780 according to your policy of failover.
881 781
882 panic_on_oom=2+kdump gives you very strong too 782 panic_on_oom=2+kdump gives you very strong tool to investigate
883 why oom happens. You can get snapshot. 783 why oom happens. You can get snapshot.
884 784
885 785
886 percpu_pagelist_high_fraction 786 percpu_pagelist_high_fraction
887 ============================= 787 =============================
888 788
889 This is the fraction of pages in each zone tha 789 This is the fraction of pages in each zone that are can be stored to
890 per-cpu page lists. It is an upper boundary th 790 per-cpu page lists. It is an upper boundary that is divided depending
891 on the number of online CPUs. The min value fo 791 on the number of online CPUs. The min value for this is 8 which means
892 that we do not allow more than 1/8th of pages 792 that we do not allow more than 1/8th of pages in each zone to be stored
893 on per-cpu page lists. This entry only changes 793 on per-cpu page lists. This entry only changes the value of hot per-cpu
894 page lists. A user can specify a number like 1 794 page lists. A user can specify a number like 100 to allocate 1/100th of
895 each zone between per-cpu lists. 795 each zone between per-cpu lists.
896 796
897 The batch value of each per-cpu page list rema 797 The batch value of each per-cpu page list remains the same regardless of
898 the value of the high fraction so allocation l 798 the value of the high fraction so allocation latencies are unaffected.
899 799
900 The initial value is zero. Kernel uses this va 800 The initial value is zero. Kernel uses this value to set the high pcp->high
901 mark based on the low watermark for the zone a 801 mark based on the low watermark for the zone and the number of local
902 online CPUs. If the user writes '0' to this s 802 online CPUs. If the user writes '0' to this sysctl, it will revert to
903 this default behavior. 803 this default behavior.
904 804
905 805
906 stat_interval 806 stat_interval
907 ============= 807 =============
908 808
909 The time interval between which vm statistics 809 The time interval between which vm statistics are updated. The default
910 is 1 second. 810 is 1 second.
911 811
912 812
913 stat_refresh 813 stat_refresh
914 ============ 814 ============
915 815
916 Any read or write (by root only) flushes all t 816 Any read or write (by root only) flushes all the per-cpu vm statistics
917 into their global totals, for more accurate re 817 into their global totals, for more accurate reports when testing
918 e.g. cat /proc/sys/vm/stat_refresh /proc/memin 818 e.g. cat /proc/sys/vm/stat_refresh /proc/meminfo
919 819
920 As a side-effect, it also checks for negative 820 As a side-effect, it also checks for negative totals (elsewhere reported
921 as 0) and "fails" with EINVAL if any are found 821 as 0) and "fails" with EINVAL if any are found, with a warning in dmesg.
922 (At time of writing, a few stats are known som 822 (At time of writing, a few stats are known sometimes to be found negative,
923 with no ill effects: errors and warnings on th 823 with no ill effects: errors and warnings on these stats are suppressed.)
924 824
925 825
926 numa_stat 826 numa_stat
927 ========= 827 =========
928 828
929 This interface allows runtime configuration of 829 This interface allows runtime configuration of numa statistics.
930 830
931 When page allocation performance becomes a bot 831 When page allocation performance becomes a bottleneck and you can tolerate
932 some possible tool breakage and decreased numa 832 some possible tool breakage and decreased numa counter precision, you can
933 do:: 833 do::
934 834
935 echo 0 > /proc/sys/vm/numa_stat 835 echo 0 > /proc/sys/vm/numa_stat
936 836
937 When page allocation performance is not a bott 837 When page allocation performance is not a bottleneck and you want all
938 tooling to work, you can do:: 838 tooling to work, you can do::
939 839
940 echo 1 > /proc/sys/vm/numa_stat 840 echo 1 > /proc/sys/vm/numa_stat
941 841
942 842
943 swappiness 843 swappiness
944 ========== 844 ==========
945 845
946 This control is used to define the rough relat 846 This control is used to define the rough relative IO cost of swapping
947 and filesystem paging, as a value between 0 an 847 and filesystem paging, as a value between 0 and 200. At 100, the VM
948 assumes equal IO cost and will thus apply memo 848 assumes equal IO cost and will thus apply memory pressure to the page
949 cache and swap-backed pages equally; lower val 849 cache and swap-backed pages equally; lower values signify more
950 expensive swap IO, higher values indicates che 850 expensive swap IO, higher values indicates cheaper.
951 851
952 Keep in mind that filesystem IO patterns under 852 Keep in mind that filesystem IO patterns under memory pressure tend to
953 be more efficient than swap's random IO. An op 853 be more efficient than swap's random IO. An optimal value will require
954 experimentation and will also be workload-depe 854 experimentation and will also be workload-dependent.
955 855
956 The default value is 60. 856 The default value is 60.
957 857
958 For in-memory swap, like zram or zswap, as wel 858 For in-memory swap, like zram or zswap, as well as hybrid setups that
959 have swap on faster devices than the filesyste 859 have swap on faster devices than the filesystem, values beyond 100 can
960 be considered. For example, if the random IO a 860 be considered. For example, if the random IO against the swap device
961 is on average 2x faster than IO from the files 861 is on average 2x faster than IO from the filesystem, swappiness should
962 be 133 (x + 2x = 200, 2x = 133.33). 862 be 133 (x + 2x = 200, 2x = 133.33).
963 863
964 At 0, the kernel will not initiate swap until 864 At 0, the kernel will not initiate swap until the amount of free and
965 file-backed pages is less than the high waterm 865 file-backed pages is less than the high watermark in a zone.
966 866
967 867
968 unprivileged_userfaultfd 868 unprivileged_userfaultfd
969 ======================== 869 ========================
970 870
971 This flag controls the mode in which unprivile 871 This flag controls the mode in which unprivileged users can use the
972 userfaultfd system calls. Set this to 0 to res 872 userfaultfd system calls. Set this to 0 to restrict unprivileged users
973 to handle page faults in user mode only. In th 873 to handle page faults in user mode only. In this case, users without
974 SYS_CAP_PTRACE must pass UFFD_USER_MODE_ONLY i 874 SYS_CAP_PTRACE must pass UFFD_USER_MODE_ONLY in order for userfaultfd to
975 succeed. Prohibiting use of userfaultfd for ha 875 succeed. Prohibiting use of userfaultfd for handling faults from kernel
976 mode may make certain vulnerabilities more dif 876 mode may make certain vulnerabilities more difficult to exploit.
977 877
978 Set this to 1 to allow unprivileged users to u 878 Set this to 1 to allow unprivileged users to use the userfaultfd system
979 calls without any restrictions. 879 calls without any restrictions.
980 880
981 The default value is 0. 881 The default value is 0.
982 882
983 Another way to control permissions for userfau <<
984 /dev/userfaultfd instead of userfaultfd(2). Se <<
985 Documentation/admin-guide/mm/userfaultfd.rst. <<
986 883
987 user_reserve_kbytes 884 user_reserve_kbytes
988 =================== 885 ===================
989 886
990 When overcommit_memory is set to 2, "never ove 887 When overcommit_memory is set to 2, "never overcommit" mode, reserve
991 min(3% of current process size, user_reserve_k 888 min(3% of current process size, user_reserve_kbytes) of free memory.
992 This is intended to prevent a user from starti 889 This is intended to prevent a user from starting a single memory hogging
993 process, such that they cannot recover (kill t 890 process, such that they cannot recover (kill the hog).
994 891
995 user_reserve_kbytes defaults to min(3% of the 892 user_reserve_kbytes defaults to min(3% of the current process size, 128MB).
996 893
997 If this is reduced to zero, then the user will 894 If this is reduced to zero, then the user will be allowed to allocate
998 all free memory with a single process, minus a 895 all free memory with a single process, minus admin_reserve_kbytes.
999 Any subsequent attempts to execute a command w 896 Any subsequent attempts to execute a command will result in
1000 "fork: Cannot allocate memory". 897 "fork: Cannot allocate memory".
1001 898
1002 Changing this takes effect whenever an applic 899 Changing this takes effect whenever an application requests memory.
1003 900
1004 901
1005 vfs_cache_pressure 902 vfs_cache_pressure
1006 ================== 903 ==================
1007 904
1008 This percentage value controls the tendency o 905 This percentage value controls the tendency of the kernel to reclaim
1009 the memory which is used for caching of direc 906 the memory which is used for caching of directory and inode objects.
1010 907
1011 At the default value of vfs_cache_pressure=10 908 At the default value of vfs_cache_pressure=100 the kernel will attempt to
1012 reclaim dentries and inodes at a "fair" rate 909 reclaim dentries and inodes at a "fair" rate with respect to pagecache and
1013 swapcache reclaim. Decreasing vfs_cache_pres 910 swapcache reclaim. Decreasing vfs_cache_pressure causes the kernel to prefer
1014 to retain dentry and inode caches. When vfs_c 911 to retain dentry and inode caches. When vfs_cache_pressure=0, the kernel will
1015 never reclaim dentries and inodes due to memo 912 never reclaim dentries and inodes due to memory pressure and this can easily
1016 lead to out-of-memory conditions. Increasing 913 lead to out-of-memory conditions. Increasing vfs_cache_pressure beyond 100
1017 causes the kernel to prefer to reclaim dentri 914 causes the kernel to prefer to reclaim dentries and inodes.
1018 915
1019 Increasing vfs_cache_pressure significantly b 916 Increasing vfs_cache_pressure significantly beyond 100 may have negative
1020 performance impact. Reclaim code needs to tak 917 performance impact. Reclaim code needs to take various locks to find freeable
1021 directory and inode objects. With vfs_cache_p 918 directory and inode objects. With vfs_cache_pressure=1000, it will look for
1022 ten times more freeable objects than there ar 919 ten times more freeable objects than there are.
1023 920
1024 921
1025 watermark_boost_factor 922 watermark_boost_factor
1026 ====================== 923 ======================
1027 924
1028 This factor controls the level of reclaim whe 925 This factor controls the level of reclaim when memory is being fragmented.
1029 It defines the percentage of the high waterma 926 It defines the percentage of the high watermark of a zone that will be
1030 reclaimed if pages of different mobility are 927 reclaimed if pages of different mobility are being mixed within pageblocks.
1031 The intent is that compaction has less work t 928 The intent is that compaction has less work to do in the future and to
1032 increase the success rate of future high-orde 929 increase the success rate of future high-order allocations such as SLUB
1033 allocations, THP and hugetlbfs pages. 930 allocations, THP and hugetlbfs pages.
1034 931
1035 To make it sensible with respect to the water 932 To make it sensible with respect to the watermark_scale_factor
1036 parameter, the unit is in fractions of 10,000 933 parameter, the unit is in fractions of 10,000. The default value of
1037 15,000 means that up to 150% of the high wate 934 15,000 means that up to 150% of the high watermark will be reclaimed in the
1038 event of a pageblock being mixed due to fragm 935 event of a pageblock being mixed due to fragmentation. The level of reclaim
1039 is determined by the number of fragmentation 936 is determined by the number of fragmentation events that occurred in the
1040 recent past. If this value is smaller than a 937 recent past. If this value is smaller than a pageblock then a pageblocks
1041 worth of pages will be reclaimed (e.g. 2MB o 938 worth of pages will be reclaimed (e.g. 2MB on 64-bit x86). A boost factor
1042 of 0 will disable the feature. 939 of 0 will disable the feature.
1043 940
1044 941
1045 watermark_scale_factor 942 watermark_scale_factor
1046 ====================== 943 ======================
1047 944
1048 This factor controls the aggressiveness of ks 945 This factor controls the aggressiveness of kswapd. It defines the
1049 amount of memory left in a node/system before 946 amount of memory left in a node/system before kswapd is woken up and
1050 how much memory needs to be free before kswap 947 how much memory needs to be free before kswapd goes back to sleep.
1051 948
1052 The unit is in fractions of 10,000. The defau 949 The unit is in fractions of 10,000. The default value of 10 means the
1053 distances between watermarks are 0.1% of the 950 distances between watermarks are 0.1% of the available memory in the
1054 node/system. The maximum value is 3000, or 30 951 node/system. The maximum value is 3000, or 30% of memory.
1055 952
1056 A high rate of threads entering direct reclai 953 A high rate of threads entering direct reclaim (allocstall) or kswapd
1057 going to sleep prematurely (kswapd_low_wmark_ 954 going to sleep prematurely (kswapd_low_wmark_hit_quickly) can indicate
1058 that the number of free pages kswapd maintain 955 that the number of free pages kswapd maintains for latency reasons is
1059 too small for the allocation bursts occurring 956 too small for the allocation bursts occurring in the system. This knob
1060 can then be used to tune kswapd aggressivenes 957 can then be used to tune kswapd aggressiveness accordingly.
1061 958
1062 959
1063 zone_reclaim_mode 960 zone_reclaim_mode
1064 ================= 961 =================
1065 962
1066 Zone_reclaim_mode allows someone to set more 963 Zone_reclaim_mode allows someone to set more or less aggressive approaches to
1067 reclaim memory when a zone runs out of memory 964 reclaim memory when a zone runs out of memory. If it is set to zero then no
1068 zone reclaim occurs. Allocations will be sati 965 zone reclaim occurs. Allocations will be satisfied from other zones / nodes
1069 in the system. 966 in the system.
1070 967
1071 This is value OR'ed together of 968 This is value OR'ed together of
1072 969
1073 = =================================== 970 = ===================================
1074 1 Zone reclaim on 971 1 Zone reclaim on
1075 2 Zone reclaim writes dirty pages out 972 2 Zone reclaim writes dirty pages out
1076 4 Zone reclaim swaps pages 973 4 Zone reclaim swaps pages
1077 = =================================== 974 = ===================================
1078 975
1079 zone_reclaim_mode is disabled by default. Fo 976 zone_reclaim_mode is disabled by default. For file servers or workloads
1080 that benefit from having their data cached, z 977 that benefit from having their data cached, zone_reclaim_mode should be
1081 left disabled as the caching effect is likely 978 left disabled as the caching effect is likely to be more important than
1082 data locality. 979 data locality.
1083 980
1084 Consider enabling one or more zone_reclaim mo 981 Consider enabling one or more zone_reclaim mode bits if it's known that the
1085 workload is partitioned such that each partit 982 workload is partitioned such that each partition fits within a NUMA node
1086 and that accessing remote memory would cause 983 and that accessing remote memory would cause a measurable performance
1087 reduction. The page allocator will take addi 984 reduction. The page allocator will take additional actions before
1088 allocating off node pages. 985 allocating off node pages.
1089 986
1090 Allowing zone reclaim to write out pages stop 987 Allowing zone reclaim to write out pages stops processes that are
1091 writing large amounts of data from dirtying p 988 writing large amounts of data from dirtying pages on other nodes. Zone
1092 reclaim will write out dirty pages if a zone 989 reclaim will write out dirty pages if a zone fills up and so effectively
1093 throttle the process. This may decrease the p 990 throttle the process. This may decrease the performance of a single process
1094 since it cannot use all of system memory to b 991 since it cannot use all of system memory to buffer the outgoing writes
1095 anymore but it preserve the memory on other n 992 anymore but it preserve the memory on other nodes so that the performance
1096 of other processes running on other nodes wil 993 of other processes running on other nodes will not be affected.
1097 994
1098 Allowing regular swap effectively restricts a 995 Allowing regular swap effectively restricts allocations to the local
1099 node unless explicitly overridden by memory p 996 node unless explicitly overridden by memory policies or cpuset
1100 configurations. 997 configurations.
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.