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 - block_dump
28 - compact_memory 29 - compact_memory
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 <<
42 - hugetlb_shm_group 40 - hugetlb_shm_group
43 - laptop_mode 41 - laptop_mode
44 - legacy_va_layout 42 - legacy_va_layout
45 - lowmem_reserve_ratio 43 - lowmem_reserve_ratio
46 - max_map_count 44 - max_map_count
47 - mem_profiling (only if CONFIG_MEM_AL <<
48 - memory_failure_early_kill 45 - memory_failure_early_kill
49 - memory_failure_recovery 46 - memory_failure_recovery
50 - min_free_kbytes 47 - min_free_kbytes
51 - min_slab_ratio 48 - min_slab_ratio
52 - min_unmapped_ratio 49 - min_unmapped_ratio
53 - mmap_min_addr 50 - mmap_min_addr
54 - mmap_rnd_bits 51 - mmap_rnd_bits
55 - mmap_rnd_compat_bits 52 - mmap_rnd_compat_bits
56 - nr_hugepages 53 - nr_hugepages
57 - nr_hugepages_mempolicy 54 - nr_hugepages_mempolicy
58 - nr_overcommit_hugepages 55 - nr_overcommit_hugepages
59 - nr_trim_pages (only if CONFIG_MMU=n) 56 - nr_trim_pages (only if CONFIG_MMU=n)
60 - numa_zonelist_order 57 - numa_zonelist_order
61 - oom_dump_tasks 58 - oom_dump_tasks
62 - oom_kill_allocating_task 59 - oom_kill_allocating_task
63 - overcommit_kbytes 60 - overcommit_kbytes
64 - overcommit_memory 61 - overcommit_memory
65 - overcommit_ratio 62 - overcommit_ratio
66 - page-cluster 63 - page-cluster
67 - page_lock_unfairness <<
68 - panic_on_oom 64 - panic_on_oom
69 - percpu_pagelist_high_fraction !! 65 - percpu_pagelist_fraction
70 - stat_interval 66 - stat_interval
71 - stat_refresh 67 - stat_refresh
72 - numa_stat 68 - numa_stat
73 - swappiness 69 - swappiness
74 - unprivileged_userfaultfd 70 - unprivileged_userfaultfd
75 - user_reserve_kbytes 71 - user_reserve_kbytes
76 - vfs_cache_pressure 72 - vfs_cache_pressure
77 - watermark_boost_factor 73 - watermark_boost_factor
78 - watermark_scale_factor 74 - watermark_scale_factor
79 - zone_reclaim_mode 75 - zone_reclaim_mode
80 76
81 77
82 admin_reserve_kbytes 78 admin_reserve_kbytes
83 ==================== 79 ====================
84 80
85 The amount of free memory in the system that s 81 The amount of free memory in the system that should be reserved for users
86 with the capability cap_sys_admin. 82 with the capability cap_sys_admin.
87 83
88 admin_reserve_kbytes defaults to min(3% of fre 84 admin_reserve_kbytes defaults to min(3% of free pages, 8MB)
89 85
90 That should provide enough for the admin to lo 86 That should provide enough for the admin to log in and kill a process,
91 if necessary, under the default overcommit 'gu 87 if necessary, under the default overcommit 'guess' mode.
92 88
93 Systems running under overcommit 'never' shoul 89 Systems running under overcommit 'never' should increase this to account
94 for the full Virtual Memory Size of programs u 90 for the full Virtual Memory Size of programs used to recover. Otherwise,
95 root may not be able to log in to recover the 91 root may not be able to log in to recover the system.
96 92
97 How do you calculate a minimum useful reserve? 93 How do you calculate a minimum useful reserve?
98 94
99 sshd or login + bash (or some other shell) + t 95 sshd or login + bash (or some other shell) + top (or ps, kill, etc.)
100 96
101 For overcommit 'guess', we can sum resident se 97 For overcommit 'guess', we can sum resident set sizes (RSS).
102 On x86_64 this is about 8MB. 98 On x86_64 this is about 8MB.
103 99
104 For overcommit 'never', we can take the max of 100 For overcommit 'never', we can take the max of their virtual sizes (VSZ)
105 and add the sum of their RSS. 101 and add the sum of their RSS.
106 On x86_64 this is about 128MB. 102 On x86_64 this is about 128MB.
107 103
108 Changing this takes effect whenever an applica 104 Changing this takes effect whenever an application requests memory.
109 105
110 106
>> 107 block_dump
>> 108 ==========
>> 109
>> 110 block_dump enables block I/O debugging when set to a nonzero value. More
>> 111 information on block I/O debugging is in Documentation/admin-guide/laptops/laptop-mode.rst.
>> 112
>> 113
111 compact_memory 114 compact_memory
112 ============== 115 ==============
113 116
114 Available only when CONFIG_COMPACTION is set. 117 Available only when CONFIG_COMPACTION is set. When 1 is written to the file,
115 all zones are compacted such that free memory 118 all zones are compacted such that free memory is available in contiguous
116 blocks where possible. This can be important f 119 blocks where possible. This can be important for example in the allocation of
117 huge pages although processes will also direct 120 huge pages although processes will also directly compact memory as required.
118 121
119 compaction_proactiveness <<
120 ======================== <<
121 <<
122 This tunable takes a value in the range [0, 10 <<
123 20. This tunable determines how aggressively c <<
124 background. Write of a non zero value to this <<
125 trigger the proactive compaction. Setting it t <<
126 <<
127 Note that compaction has a non-trivial system- <<
128 belonging to different processes are moved aro <<
129 to latency spikes in unsuspecting applications <<
130 various heuristics to avoid wasting CPU cycles <<
131 proactive compaction is not being effective. <<
132 <<
133 Be careful when setting it to extreme values l <<
134 cause excessive background compaction activity <<
135 122
136 compact_unevictable_allowed 123 compact_unevictable_allowed
137 =========================== 124 ===========================
138 125
139 Available only when CONFIG_COMPACTION is set. 126 Available only when CONFIG_COMPACTION is set. When set to 1, compaction is
140 allowed to examine the unevictable lru (mlocke 127 allowed to examine the unevictable lru (mlocked pages) for pages to compact.
141 This should be used on systems where stalls fo 128 This should be used on systems where stalls for minor page faults are an
142 acceptable trade for large contiguous free mem 129 acceptable trade for large contiguous free memory. Set to 0 to prevent
143 compaction from moving pages that are unevicta 130 compaction from moving pages that are unevictable. Default value is 1.
144 On CONFIG_PREEMPT_RT the default value is 0 in 131 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 !! 132 to compaction, which would block the task from becomming active until the fault
146 is resolved. 133 is resolved.
147 134
148 135
149 dirty_background_bytes 136 dirty_background_bytes
150 ====================== 137 ======================
151 138
152 Contains the amount of dirty memory at which t 139 Contains the amount of dirty memory at which the background kernel
153 flusher threads will start writeback. 140 flusher threads will start writeback.
154 141
155 Note: 142 Note:
156 dirty_background_bytes is the counterpart of 143 dirty_background_bytes is the counterpart of dirty_background_ratio. Only
157 one of them may be specified at a time. When 144 one of them may be specified at a time. When one sysctl is written it is
158 immediately taken into account to evaluate t 145 immediately taken into account to evaluate the dirty memory limits and the
159 other appears as 0 when read. 146 other appears as 0 when read.
160 147
161 148
162 dirty_background_ratio 149 dirty_background_ratio
163 ====================== 150 ======================
164 151
165 Contains, as a percentage of total available m 152 Contains, as a percentage of total available memory that contains free pages
166 and reclaimable pages, the number of pages at 153 and reclaimable pages, the number of pages at which the background kernel
167 flusher threads will start writing out dirty d 154 flusher threads will start writing out dirty data.
168 155
169 The total available memory is not equal to tot 156 The total available memory is not equal to total system memory.
170 157
171 158
172 dirty_bytes 159 dirty_bytes
173 =========== 160 ===========
174 161
175 Contains the amount of dirty memory at which a 162 Contains the amount of dirty memory at which a process generating disk writes
176 will itself start writeback. 163 will itself start writeback.
177 164
178 Note: dirty_bytes is the counterpart of dirty_ 165 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 166 specified at a time. When one sysctl is written it is immediately taken into
180 account to evaluate the dirty memory limits an 167 account to evaluate the dirty memory limits and the other appears as 0 when
181 read. 168 read.
182 169
183 Note: the minimum value allowed for dirty_byte 170 Note: the minimum value allowed for dirty_bytes is two pages (in bytes); any
184 value lower than this limit will be ignored an 171 value lower than this limit will be ignored and the old configuration will be
185 retained. 172 retained.
186 173
187 174
188 dirty_expire_centisecs 175 dirty_expire_centisecs
189 ====================== 176 ======================
190 177
191 This tunable is used to define when dirty data 178 This tunable is used to define when dirty data is old enough to be eligible
192 for writeout by the kernel flusher threads. I 179 for writeout by the kernel flusher threads. It is expressed in 100'ths
193 of a second. Data which has been dirty in-mem 180 of a second. Data which has been dirty in-memory for longer than this
194 interval will be written out next time a flush 181 interval will be written out next time a flusher thread wakes up.
195 182
196 183
197 dirty_ratio 184 dirty_ratio
198 =========== 185 ===========
199 186
200 Contains, as a percentage of total available m 187 Contains, as a percentage of total available memory that contains free pages
201 and reclaimable pages, the number of pages at 188 and reclaimable pages, the number of pages at which a process which is
202 generating disk writes will itself start writi 189 generating disk writes will itself start writing out dirty data.
203 190
204 The total available memory is not equal to tot 191 The total available memory is not equal to total system memory.
205 192
206 193
207 dirtytime_expire_seconds 194 dirtytime_expire_seconds
208 ======================== 195 ========================
209 196
210 When a lazytime inode is constantly having its 197 When a lazytime inode is constantly having its pages dirtied, the inode with
211 an updated timestamp will never get chance to 198 an updated timestamp will never get chance to be written out. And, if the
212 only thing that has happened on the file syste 199 only thing that has happened on the file system is a dirtytime inode caused
213 by an atime update, a worker will be scheduled 200 by an atime update, a worker will be scheduled to make sure that inode
214 eventually gets pushed out to disk. This tuna 201 eventually gets pushed out to disk. This tunable is used to define when dirty
215 inode is old enough to be eligible for writeba 202 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 203 And, it is also used as the interval to wakeup dirtytime_writeback thread.
217 204
218 205
219 dirty_writeback_centisecs 206 dirty_writeback_centisecs
220 ========================= 207 =========================
221 208
222 The kernel flusher threads will periodically w 209 The kernel flusher threads will periodically wake up and write `old` data
223 out to disk. This tunable expresses the inter 210 out to disk. This tunable expresses the interval between those wakeups, in
224 100'ths of a second. 211 100'ths of a second.
225 212
226 Setting this to zero disables periodic writeba 213 Setting this to zero disables periodic writeback altogether.
227 214
228 215
229 drop_caches 216 drop_caches
230 =========== 217 ===========
231 218
232 Writing to this will cause the kernel to drop 219 Writing to this will cause the kernel to drop clean caches, as well as
233 reclaimable slab objects like dentries and ino 220 reclaimable slab objects like dentries and inodes. Once dropped, their
234 memory becomes free. 221 memory becomes free.
235 222
236 To free pagecache:: 223 To free pagecache::
237 224
238 echo 1 > /proc/sys/vm/drop_caches 225 echo 1 > /proc/sys/vm/drop_caches
239 226
240 To free reclaimable slab objects (includes den 227 To free reclaimable slab objects (includes dentries and inodes)::
241 228
242 echo 2 > /proc/sys/vm/drop_caches 229 echo 2 > /proc/sys/vm/drop_caches
243 230
244 To free slab objects and pagecache:: 231 To free slab objects and pagecache::
245 232
246 echo 3 > /proc/sys/vm/drop_caches 233 echo 3 > /proc/sys/vm/drop_caches
247 234
248 This is a non-destructive operation and will n 235 This is a non-destructive operation and will not free any dirty objects.
249 To increase the number of objects freed by thi 236 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 237 `sync` prior to writing to /proc/sys/vm/drop_caches. This will minimize the
251 number of dirty objects on the system and crea 238 number of dirty objects on the system and create more candidates to be
252 dropped. 239 dropped.
253 240
254 This file is not a means to control the growth 241 This file is not a means to control the growth of the various kernel caches
255 (inodes, dentries, pagecache, etc...) These o 242 (inodes, dentries, pagecache, etc...) These objects are automatically
256 reclaimed by the kernel when memory is needed 243 reclaimed by the kernel when memory is needed elsewhere on the system.
257 244
258 Use of this file can cause performance problem 245 Use of this file can cause performance problems. Since it discards cached
259 objects, it may cost a significant amount of I 246 objects, it may cost a significant amount of I/O and CPU to recreate the
260 dropped objects, especially if they were under 247 dropped objects, especially if they were under heavy use. Because of this,
261 use outside of a testing or debugging environm 248 use outside of a testing or debugging environment is not recommended.
262 249
263 You may see informational messages in your ker 250 You may see informational messages in your kernel log when this file is
264 used:: 251 used::
265 252
266 cat (1234): drop_caches: 3 253 cat (1234): drop_caches: 3
267 254
268 These are informational only. They do not mea 255 These are informational only. They do not mean that anything is wrong
269 with your system. To disable them, echo 4 (bi 256 with your system. To disable them, echo 4 (bit 2) into drop_caches.
270 257
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 <<
308 258
309 extfrag_threshold 259 extfrag_threshold
310 ================= 260 =================
311 261
312 This parameter affects whether the kernel will 262 This parameter affects whether the kernel will compact memory or direct
313 reclaim to satisfy a high-order allocation. Th 263 reclaim to satisfy a high-order allocation. The extfrag/extfrag_index file in
314 debugfs shows what the fragmentation index for 264 debugfs shows what the fragmentation index for each order is in each zone in
315 the system. Values tending towards 0 imply all 265 the system. Values tending towards 0 imply allocations would fail due to lack
316 of memory, values towards 1000 imply failures 266 of memory, values towards 1000 imply failures are due to fragmentation and -1
317 implies that the allocation will succeed as lo 267 implies that the allocation will succeed as long as watermarks are met.
318 268
319 The kernel will not compact memory in a zone i 269 The kernel will not compact memory in a zone if the
320 fragmentation index is <= extfrag_threshold. T 270 fragmentation index is <= extfrag_threshold. The default value is 500.
321 271
322 272
323 highmem_is_dirtyable 273 highmem_is_dirtyable
324 ==================== 274 ====================
325 275
326 Available only for systems with CONFIG_HIGHMEM 276 Available only for systems with CONFIG_HIGHMEM enabled (32b systems).
327 277
328 This parameter controls whether the high memor 278 This parameter controls whether the high memory is considered for dirty
329 writers throttling. This is not the case by d 279 writers throttling. This is not the case by default which means that
330 only the amount of memory directly visible/usa 280 only the amount of memory directly visible/usable by the kernel can
331 be dirtied. As a result, on systems with a lar 281 be dirtied. As a result, on systems with a large amount of memory and
332 lowmem basically depleted writers might be thr 282 lowmem basically depleted writers might be throttled too early and
333 streaming writes can get very slow. 283 streaming writes can get very slow.
334 284
335 Changing the value to non zero would allow mor 285 Changing the value to non zero would allow more memory to be dirtied
336 and thus allow writers to write more data whic 286 and thus allow writers to write more data which can be flushed to the
337 storage more effectively. Note this also comes 287 storage more effectively. Note this also comes with a risk of pre-mature
338 OOM killer because some writers (e.g. direct b 288 OOM killer because some writers (e.g. direct block device writes) can
339 only use the low memory and they can fill it u 289 only use the low memory and they can fill it up with dirty data without
340 any throttling. 290 any throttling.
341 291
342 292
343 hugetlb_shm_group 293 hugetlb_shm_group
344 ================= 294 =================
345 295
346 hugetlb_shm_group contains group id that is al 296 hugetlb_shm_group contains group id that is allowed to create SysV
347 shared memory segment using hugetlb page. 297 shared memory segment using hugetlb page.
348 298
349 299
350 laptop_mode 300 laptop_mode
351 =========== 301 ===========
352 302
353 laptop_mode is a knob that controls "laptop mo 303 laptop_mode is a knob that controls "laptop mode". All the things that are
354 controlled by this knob are discussed in Docum 304 controlled by this knob are discussed in Documentation/admin-guide/laptops/laptop-mode.rst.
355 305
356 306
357 legacy_va_layout 307 legacy_va_layout
358 ================ 308 ================
359 309
360 If non-zero, this sysctl disables the new 32-b 310 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 311 will use the legacy (2.4) layout for all processes.
362 312
363 313
364 lowmem_reserve_ratio 314 lowmem_reserve_ratio
365 ==================== 315 ====================
366 316
367 For some specialised workloads on highmem mach 317 For some specialised workloads on highmem machines it is dangerous for
368 the kernel to allow process memory to be alloc 318 the kernel to allow process memory to be allocated from the "lowmem"
369 zone. This is because that memory could then 319 zone. This is because that memory could then be pinned via the mlock()
370 system call, or by unavailability of swapspace 320 system call, or by unavailability of swapspace.
371 321
372 And on large highmem machines this lack of rec 322 And on large highmem machines this lack of reclaimable lowmem memory
373 can be fatal. 323 can be fatal.
374 324
375 So the Linux page allocator has a mechanism wh 325 So the Linux page allocator has a mechanism which prevents allocations
376 which *could* use highmem from using too much 326 which *could* use highmem from using too much lowmem. This means that
377 a certain amount of lowmem is defended from th 327 a certain amount of lowmem is defended from the possibility of being
378 captured into pinned user memory. 328 captured into pinned user memory.
379 329
380 (The same argument applies to the old 16 megab 330 (The same argument applies to the old 16 megabyte ISA DMA region. This
381 mechanism will also defend that region from al 331 mechanism will also defend that region from allocations which could use
382 highmem or lowmem). 332 highmem or lowmem).
383 333
384 The `lowmem_reserve_ratio` tunable determines 334 The `lowmem_reserve_ratio` tunable determines how aggressive the kernel is
385 in defending these lower zones. 335 in defending these lower zones.
386 336
387 If you have a machine which uses highmem or IS 337 If you have a machine which uses highmem or ISA DMA and your
388 applications are using mlock(), or if you are 338 applications are using mlock(), or if you are running with no swap then
389 you probably should change the lowmem_reserve_ 339 you probably should change the lowmem_reserve_ratio setting.
390 340
391 The lowmem_reserve_ratio is an array. You can 341 The lowmem_reserve_ratio is an array. You can see them by reading this file::
392 342
393 % cat /proc/sys/vm/lowmem_reserve_rati 343 % cat /proc/sys/vm/lowmem_reserve_ratio
394 256 256 32 344 256 256 32
395 345
396 But, these values are not used directly. The k 346 But, these values are not used directly. The kernel calculates # of protection
397 pages for each zones from them. These are show 347 pages for each zones from them. These are shown as array of protection pages
398 in /proc/zoneinfo like the following. (This is !! 348 in /proc/zoneinfo like followings. (This is an example of x86-64 box).
399 Each zone has an array of protection pages lik 349 Each zone has an array of protection pages like this::
400 350
401 Node 0, zone DMA 351 Node 0, zone DMA
402 pages free 1355 352 pages free 1355
403 min 3 353 min 3
404 low 3 354 low 3
405 high 4 355 high 4
406 : 356 :
407 : 357 :
408 numa_other 0 358 numa_other 0
409 protection: (0, 2004, 2004, 2004) 359 protection: (0, 2004, 2004, 2004)
410 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 360 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
411 pagesets 361 pagesets
412 cpu: 0 pcp: 0 362 cpu: 0 pcp: 0
413 : 363 :
414 364
415 These protections are added to score to judge 365 These protections are added to score to judge whether this zone should be used
416 for page allocation or should be reclaimed. 366 for page allocation or should be reclaimed.
417 367
418 In this example, if normal pages (index=2) are 368 In this example, if normal pages (index=2) are required to this DMA zone and
419 watermark[WMARK_HIGH] is used for watermark, t 369 watermark[WMARK_HIGH] is used for watermark, the kernel judges this zone should
420 not be used because pages_free(1355) is smalle 370 not be used because pages_free(1355) is smaller than watermark + protection[2]
421 (4 + 2004 = 2008). If this protection value is 371 (4 + 2004 = 2008). If this protection value is 0, this zone would be used for
422 normal page requirement. If requirement is DMA 372 normal page requirement. If requirement is DMA zone(index=0), protection[0]
423 (=0) is used. 373 (=0) is used.
424 374
425 zone[i]'s protection[j] is calculated by follo 375 zone[i]'s protection[j] is calculated by following expression::
426 376
427 (i < j): 377 (i < j):
428 zone[i]->protection[j] 378 zone[i]->protection[j]
429 = (total sums of managed_pages from zone[i 379 = (total sums of managed_pages from zone[i+1] to zone[j] on the node)
430 / lowmem_reserve_ratio[i]; 380 / lowmem_reserve_ratio[i];
431 (i = j): 381 (i = j):
432 (should not be protected. = 0; 382 (should not be protected. = 0;
433 (i > j): 383 (i > j):
434 (not necessary, but looks 0) 384 (not necessary, but looks 0)
435 385
436 The default values of lowmem_reserve_ratio[i] 386 The default values of lowmem_reserve_ratio[i] are
437 387
438 === ==================================== 388 === ====================================
439 256 (if zone[i] means DMA or DMA32 zone) 389 256 (if zone[i] means DMA or DMA32 zone)
440 32 (others) 390 32 (others)
441 === ==================================== 391 === ====================================
442 392
443 As above expression, they are reciprocal numbe 393 As above expression, they are reciprocal number of ratio.
444 256 means 1/256. # of protection pages becomes 394 256 means 1/256. # of protection pages becomes about "0.39%" of total managed
445 pages of higher zones on the node. 395 pages of higher zones on the node.
446 396
447 If you would like to protect more pages, small 397 If you would like to protect more pages, smaller values are effective.
448 The minimum value is 1 (1/1 -> 100%). The valu 398 The minimum value is 1 (1/1 -> 100%). The value less than 1 completely
449 disables protection of the pages. 399 disables protection of the pages.
450 400
451 401
452 max_map_count: 402 max_map_count:
453 ============== 403 ==============
454 404
455 This file contains the maximum number of memor 405 This file contains the maximum number of memory map areas a process
456 may have. Memory map areas are used as a side- 406 may have. Memory map areas are used as a side-effect of calling
457 malloc, directly by mmap, mprotect, and madvis 407 malloc, directly by mmap, mprotect, and madvise, and also when loading
458 shared libraries. 408 shared libraries.
459 409
460 While most applications need less than a thous 410 While most applications need less than a thousand maps, certain
461 programs, particularly malloc debuggers, may c 411 programs, particularly malloc debuggers, may consume lots of them,
462 e.g., up to one or two maps per allocation. 412 e.g., up to one or two maps per allocation.
463 413
464 The default value is 65530. !! 414 The default value is 65536.
465 <<
466 <<
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_ <<
480 415
481 416
482 memory_failure_early_kill: 417 memory_failure_early_kill:
483 ========================== 418 ==========================
484 419
485 Control how to kill processes when uncorrected 420 Control how to kill processes when uncorrected memory error (typically
486 a 2bit error in a memory module) is detected i 421 a 2bit error in a memory module) is detected in the background by hardware
487 that cannot be handled by the kernel. In some 422 that cannot be handled by the kernel. In some cases (like the page
488 still having a valid copy on disk) the kernel 423 still having a valid copy on disk) the kernel will handle the failure
489 transparently without affecting any applicatio 424 transparently without affecting any applications. But if there is
490 no other up-to-date copy of the data it will k !! 425 no other uptodate copy of the data it will kill to prevent any data
491 corruptions from propagating. 426 corruptions from propagating.
492 427
493 1: Kill all processes that have the corrupted 428 1: Kill all processes that have the corrupted and not reloadable page mapped
494 as soon as the corruption is detected. Note t 429 as soon as the corruption is detected. Note this is not supported
495 for a few types of pages, like kernel internal 430 for a few types of pages, like kernel internally allocated data or
496 the swap cache, but works for the majority of 431 the swap cache, but works for the majority of user pages.
497 432
498 0: Only unmap the corrupted page from all proc 433 0: Only unmap the corrupted page from all processes and only kill a process
499 who tries to access it. 434 who tries to access it.
500 435
501 The kill is done using a catchable SIGBUS with 436 The kill is done using a catchable SIGBUS with BUS_MCEERR_AO, so processes can
502 handle this if they want to. 437 handle this if they want to.
503 438
504 This is only active on architectures/platforms 439 This is only active on architectures/platforms with advanced machine
505 check handling and depends on the hardware cap 440 check handling and depends on the hardware capabilities.
506 441
507 Applications can override this setting individ 442 Applications can override this setting individually with the PR_MCE_KILL prctl
508 443
509 444
510 memory_failure_recovery 445 memory_failure_recovery
511 ======================= 446 =======================
512 447
513 Enable memory failure recovery (when supported 448 Enable memory failure recovery (when supported by the platform)
514 449
515 1: Attempt recovery. 450 1: Attempt recovery.
516 451
517 0: Always panic on a memory failure. 452 0: Always panic on a memory failure.
518 453
519 454
520 min_free_kbytes 455 min_free_kbytes
521 =============== 456 ===============
522 457
523 This is used to force the Linux VM to keep a m 458 This is used to force the Linux VM to keep a minimum number
524 of kilobytes free. The VM uses this number to 459 of kilobytes free. The VM uses this number to compute a
525 watermark[WMARK_MIN] value for each lowmem zon 460 watermark[WMARK_MIN] value for each lowmem zone in the system.
526 Each lowmem zone gets a number of reserved fre 461 Each lowmem zone gets a number of reserved free pages based
527 proportionally on its size. 462 proportionally on its size.
528 463
529 Some minimal amount of memory is needed to sat 464 Some minimal amount of memory is needed to satisfy PF_MEMALLOC
530 allocations; if you set this to lower than 102 465 allocations; if you set this to lower than 1024KB, your system will
531 become subtly broken, and prone to deadlock un 466 become subtly broken, and prone to deadlock under high loads.
532 467
533 Setting this too high will OOM your machine in 468 Setting this too high will OOM your machine instantly.
534 469
535 470
536 min_slab_ratio 471 min_slab_ratio
537 ============== 472 ==============
538 473
539 This is available only on NUMA kernels. 474 This is available only on NUMA kernels.
540 475
541 A percentage of the total pages in each zone. 476 A percentage of the total pages in each zone. On Zone reclaim
542 (fallback from the local zone occurs) slabs wi 477 (fallback from the local zone occurs) slabs will be reclaimed if more
543 than this percentage of pages in a zone are re 478 than this percentage of pages in a zone are reclaimable slab pages.
544 This insures that the slab growth stays under 479 This insures that the slab growth stays under control even in NUMA
545 systems that rarely perform global reclaim. 480 systems that rarely perform global reclaim.
546 481
547 The default is 5 percent. 482 The default is 5 percent.
548 483
549 Note that slab reclaim is triggered in a per z 484 Note that slab reclaim is triggered in a per zone / node fashion.
550 The process of reclaiming slab memory is curre 485 The process of reclaiming slab memory is currently not node specific
551 and may not be fast. 486 and may not be fast.
552 487
553 488
554 min_unmapped_ratio 489 min_unmapped_ratio
555 ================== 490 ==================
556 491
557 This is available only on NUMA kernels. 492 This is available only on NUMA kernels.
558 493
559 This is a percentage of the total pages in eac 494 This is a percentage of the total pages in each zone. Zone reclaim will
560 only occur if more than this percentage of pag 495 only occur if more than this percentage of pages are in a state that
561 zone_reclaim_mode allows to be reclaimed. 496 zone_reclaim_mode allows to be reclaimed.
562 497
563 If zone_reclaim_mode has the value 4 OR'd, the 498 If zone_reclaim_mode has the value 4 OR'd, then the percentage is compared
564 against all file-backed unmapped pages includi 499 against all file-backed unmapped pages including swapcache pages and tmpfs
565 files. Otherwise, only unmapped pages backed b 500 files. Otherwise, only unmapped pages backed by normal files but not tmpfs
566 files and similar are considered. 501 files and similar are considered.
567 502
568 The default is 1 percent. 503 The default is 1 percent.
569 504
570 505
571 mmap_min_addr 506 mmap_min_addr
572 ============= 507 =============
573 508
574 This file indicates the amount of address spac 509 This file indicates the amount of address space which a user process will
575 be restricted from mmapping. Since kernel nul 510 be restricted from mmapping. Since kernel null dereference bugs could
576 accidentally operate based on the information 511 accidentally operate based on the information in the first couple of pages
577 of memory userspace processes should not be al 512 of memory userspace processes should not be allowed to write to them. By
578 default this value is set to 0 and no protecti 513 default this value is set to 0 and no protections will be enforced by the
579 security module. Setting this value to someth 514 security module. Setting this value to something like 64k will allow the
580 vast majority of applications to work correctl 515 vast majority of applications to work correctly and provide defense in depth
581 against future potential kernel bugs. 516 against future potential kernel bugs.
582 517
583 518
584 mmap_rnd_bits 519 mmap_rnd_bits
585 ============= 520 =============
586 521
587 This value can be used to select the number of 522 This value can be used to select the number of bits to use to
588 determine the random offset to the base addres 523 determine the random offset to the base address of vma regions
589 resulting from mmap allocations on architectur 524 resulting from mmap allocations on architectures which support
590 tuning address space randomization. This valu 525 tuning address space randomization. This value will be bounded
591 by the architecture's minimum and maximum supp 526 by the architecture's minimum and maximum supported values.
592 527
593 This value can be changed after boot using the 528 This value can be changed after boot using the
594 /proc/sys/vm/mmap_rnd_bits tunable 529 /proc/sys/vm/mmap_rnd_bits tunable
595 530
596 531
597 mmap_rnd_compat_bits 532 mmap_rnd_compat_bits
598 ==================== 533 ====================
599 534
600 This value can be used to select the number of 535 This value can be used to select the number of bits to use to
601 determine the random offset to the base addres 536 determine the random offset to the base address of vma regions
602 resulting from mmap allocations for applicatio 537 resulting from mmap allocations for applications run in
603 compatibility mode on architectures which supp 538 compatibility mode on architectures which support tuning address
604 space randomization. This value will be bound 539 space randomization. This value will be bounded by the
605 architecture's minimum and maximum supported v 540 architecture's minimum and maximum supported values.
606 541
607 This value can be changed after boot using the 542 This value can be changed after boot using the
608 /proc/sys/vm/mmap_rnd_compat_bits tunable 543 /proc/sys/vm/mmap_rnd_compat_bits tunable
609 544
610 545
611 nr_hugepages 546 nr_hugepages
612 ============ 547 ============
613 548
614 Change the minimum size of the hugepage pool. 549 Change the minimum size of the hugepage pool.
615 550
616 See Documentation/admin-guide/mm/hugetlbpage.r 551 See Documentation/admin-guide/mm/hugetlbpage.rst
617 552
618 553
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 554 nr_hugepages_mempolicy
657 ====================== 555 ======================
658 556
659 Change the size of the hugepage pool at run-ti 557 Change the size of the hugepage pool at run-time on a specific
660 set of NUMA nodes. 558 set of NUMA nodes.
661 559
662 See Documentation/admin-guide/mm/hugetlbpage.r 560 See Documentation/admin-guide/mm/hugetlbpage.rst
663 561
664 562
665 nr_overcommit_hugepages 563 nr_overcommit_hugepages
666 ======================= 564 =======================
667 565
668 Change the maximum size of the hugepage pool. 566 Change the maximum size of the hugepage pool. The maximum is
669 nr_hugepages + nr_overcommit_hugepages. 567 nr_hugepages + nr_overcommit_hugepages.
670 568
671 See Documentation/admin-guide/mm/hugetlbpage.r 569 See Documentation/admin-guide/mm/hugetlbpage.rst
672 570
673 571
674 nr_trim_pages 572 nr_trim_pages
675 ============= 573 =============
676 574
677 This is available only on NOMMU kernels. 575 This is available only on NOMMU kernels.
678 576
679 This value adjusts the excess page trimming be 577 This value adjusts the excess page trimming behaviour of power-of-2 aligned
680 NOMMU mmap allocations. 578 NOMMU mmap allocations.
681 579
682 A value of 0 disables trimming of allocations 580 A value of 0 disables trimming of allocations entirely, while a value of 1
683 trims excess pages aggressively. Any value >= 581 trims excess pages aggressively. Any value >= 1 acts as the watermark where
684 trimming of allocations is initiated. 582 trimming of allocations is initiated.
685 583
686 The default value is 1. 584 The default value is 1.
687 585
688 See Documentation/admin-guide/mm/nommu-mmap.rs !! 586 See Documentation/nommu-mmap.txt for more information.
689 587
690 588
691 numa_zonelist_order 589 numa_zonelist_order
692 =================== 590 ===================
693 591
694 This sysctl is only for NUMA and it is depreca 592 This sysctl is only for NUMA and it is deprecated. Anything but
695 Node order will fail! 593 Node order will fail!
696 594
697 'where the memory is allocated from' is contro 595 'where the memory is allocated from' is controlled by zonelists.
698 596
699 (This documentation ignores ZONE_HIGHMEM/ZONE_ 597 (This documentation ignores ZONE_HIGHMEM/ZONE_DMA32 for simple explanation.
700 you may be able to read ZONE_DMA as ZONE_DMA32 598 you may be able to read ZONE_DMA as ZONE_DMA32...)
701 599
702 In non-NUMA case, a zonelist for GFP_KERNEL is 600 In non-NUMA case, a zonelist for GFP_KERNEL is ordered as following.
703 ZONE_NORMAL -> ZONE_DMA 601 ZONE_NORMAL -> ZONE_DMA
704 This means that a memory allocation request fo 602 This means that a memory allocation request for GFP_KERNEL will
705 get memory from ZONE_DMA only when ZONE_NORMAL 603 get memory from ZONE_DMA only when ZONE_NORMAL is not available.
706 604
707 In NUMA case, you can think of following 2 typ 605 In NUMA case, you can think of following 2 types of order.
708 Assume 2 node NUMA and below is zonelist of No 606 Assume 2 node NUMA and below is zonelist of Node(0)'s GFP_KERNEL::
709 607
710 (A) Node(0) ZONE_NORMAL -> Node(0) ZONE_DMA 608 (A) Node(0) ZONE_NORMAL -> Node(0) ZONE_DMA -> Node(1) ZONE_NORMAL
711 (B) Node(0) ZONE_NORMAL -> Node(1) ZONE_NORM 609 (B) Node(0) ZONE_NORMAL -> Node(1) ZONE_NORMAL -> Node(0) ZONE_DMA.
712 610
713 Type(A) offers the best locality for processes 611 Type(A) offers the best locality for processes on Node(0), but ZONE_DMA
714 will be used before ZONE_NORMAL exhaustion. Th 612 will be used before ZONE_NORMAL exhaustion. This increases possibility of
715 out-of-memory(OOM) of ZONE_DMA because ZONE_DM 613 out-of-memory(OOM) of ZONE_DMA because ZONE_DMA is tend to be small.
716 614
717 Type(B) cannot offer the best locality but is 615 Type(B) cannot offer the best locality but is more robust against OOM of
718 the DMA zone. 616 the DMA zone.
719 617
720 Type(A) is called as "Node" order. Type (B) is 618 Type(A) is called as "Node" order. Type (B) is "Zone" order.
721 619
722 "Node order" orders the zonelists by node, the 620 "Node order" orders the zonelists by node, then by zone within each node.
723 Specify "[Nn]ode" for node order 621 Specify "[Nn]ode" for node order
724 622
725 "Zone Order" orders the zonelists by zone type 623 "Zone Order" orders the zonelists by zone type, then by node within each
726 zone. Specify "[Zz]one" for zone order. 624 zone. Specify "[Zz]one" for zone order.
727 625
728 Specify "[Dd]efault" to request automatic conf 626 Specify "[Dd]efault" to request automatic configuration.
729 627
730 On 32-bit, the Normal zone needs to be preserv 628 On 32-bit, the Normal zone needs to be preserved for allocations accessible
731 by the kernel, so "zone" order will be selecte 629 by the kernel, so "zone" order will be selected.
732 630
733 On 64-bit, devices that require DMA32/DMA are 631 On 64-bit, devices that require DMA32/DMA are relatively rare, so "node"
734 order will be selected. 632 order will be selected.
735 633
736 Default order is recommended unless this is ca 634 Default order is recommended unless this is causing problems for your
737 system/application. 635 system/application.
738 636
739 637
740 oom_dump_tasks 638 oom_dump_tasks
741 ============== 639 ==============
742 640
743 Enables a system-wide task dump (excluding ker 641 Enables a system-wide task dump (excluding kernel threads) to be produced
744 when the kernel performs an OOM-killing and in 642 when the kernel performs an OOM-killing and includes such information as
745 pid, uid, tgid, vm size, rss, pgtables_bytes, 643 pid, uid, tgid, vm size, rss, pgtables_bytes, swapents, oom_score_adj
746 score, and name. This is helpful to determine 644 score, and name. This is helpful to determine why the OOM killer was
747 invoked, to identify the rogue task that cause 645 invoked, to identify the rogue task that caused it, and to determine why
748 the OOM killer chose the task it did to kill. 646 the OOM killer chose the task it did to kill.
749 647
750 If this is set to zero, this information is su 648 If this is set to zero, this information is suppressed. On very
751 large systems with thousands of tasks it may n 649 large systems with thousands of tasks it may not be feasible to dump
752 the memory state information for each one. Su 650 the memory state information for each one. Such systems should not
753 be forced to incur a performance penalty in OO 651 be forced to incur a performance penalty in OOM conditions when the
754 information may not be desired. 652 information may not be desired.
755 653
756 If this is set to non-zero, this information i 654 If this is set to non-zero, this information is shown whenever the
757 OOM killer actually kills a memory-hogging tas 655 OOM killer actually kills a memory-hogging task.
758 656
759 The default value is 1 (enabled). 657 The default value is 1 (enabled).
760 658
761 659
762 oom_kill_allocating_task 660 oom_kill_allocating_task
763 ======================== 661 ========================
764 662
765 This enables or disables killing the OOM-trigg 663 This enables or disables killing the OOM-triggering task in
766 out-of-memory situations. 664 out-of-memory situations.
767 665
768 If this is set to zero, the OOM killer will sc 666 If this is set to zero, the OOM killer will scan through the entire
769 tasklist and select a task based on heuristics 667 tasklist and select a task based on heuristics to kill. This normally
770 selects a rogue memory-hogging task that frees 668 selects a rogue memory-hogging task that frees up a large amount of
771 memory when killed. 669 memory when killed.
772 670
773 If this is set to non-zero, the OOM killer sim 671 If this is set to non-zero, the OOM killer simply kills the task that
774 triggered the out-of-memory condition. This a 672 triggered the out-of-memory condition. This avoids the expensive
775 tasklist scan. 673 tasklist scan.
776 674
777 If panic_on_oom is selected, it takes preceden 675 If panic_on_oom is selected, it takes precedence over whatever value
778 is used in oom_kill_allocating_task. 676 is used in oom_kill_allocating_task.
779 677
780 The default value is 0. 678 The default value is 0.
781 679
782 680
783 overcommit_kbytes 681 overcommit_kbytes
784 ================= 682 =================
785 683
786 When overcommit_memory is set to 2, the commit 684 When overcommit_memory is set to 2, the committed address space is not
787 permitted to exceed swap plus this amount of p 685 permitted to exceed swap plus this amount of physical RAM. See below.
788 686
789 Note: overcommit_kbytes is the counterpart of 687 Note: overcommit_kbytes is the counterpart of overcommit_ratio. Only one
790 of them may be specified at a time. Setting on 688 of them may be specified at a time. Setting one disables the other (which
791 then appears as 0 when read). 689 then appears as 0 when read).
792 690
793 691
794 overcommit_memory 692 overcommit_memory
795 ================= 693 =================
796 694
797 This value contains a flag that enables memory 695 This value contains a flag that enables memory overcommitment.
798 696
799 When this flag is 0, the kernel compares the u !! 697 When this flag is 0, the kernel attempts to estimate the amount
800 size against total memory plus swap and reject !! 698 of free memory left when userspace requests more memory.
801 699
802 When this flag is 1, the kernel pretends there 700 When this flag is 1, the kernel pretends there is always enough
803 memory until it actually runs out. 701 memory until it actually runs out.
804 702
805 When this flag is 2, the kernel uses a "never 703 When this flag is 2, the kernel uses a "never overcommit"
806 policy that attempts to prevent any overcommit 704 policy that attempts to prevent any overcommit of memory.
807 Note that user_reserve_kbytes affects this pol 705 Note that user_reserve_kbytes affects this policy.
808 706
809 This feature can be very useful because there 707 This feature can be very useful because there are a lot of
810 programs that malloc() huge amounts of memory 708 programs that malloc() huge amounts of memory "just-in-case"
811 and don't use much of it. 709 and don't use much of it.
812 710
813 The default value is 0. 711 The default value is 0.
814 712
815 See Documentation/mm/overcommit-accounting.rst !! 713 See Documentation/vm/overcommit-accounting.rst and
816 mm/util.c::__vm_enough_memory() for more infor 714 mm/util.c::__vm_enough_memory() for more information.
817 715
818 716
819 overcommit_ratio 717 overcommit_ratio
820 ================ 718 ================
821 719
822 When overcommit_memory is set to 2, the commit 720 When overcommit_memory is set to 2, the committed address
823 space is not permitted to exceed swap plus thi 721 space is not permitted to exceed swap plus this percentage
824 of physical RAM. See above. 722 of physical RAM. See above.
825 723
826 724
827 page-cluster 725 page-cluster
828 ============ 726 ============
829 727
830 page-cluster controls the number of pages up t 728 page-cluster controls the number of pages up to which consecutive pages
831 are read in from swap in a single attempt. Thi 729 are read in from swap in a single attempt. This is the swap counterpart
832 to page cache readahead. 730 to page cache readahead.
833 The mentioned consecutivity is not in terms of 731 The mentioned consecutivity is not in terms of virtual/physical addresses,
834 but consecutive on swap space - that means the 732 but consecutive on swap space - that means they were swapped out together.
835 733
836 It is a logarithmic value - setting it to zero 734 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 735 it to 1 means "2 pages", setting it to 2 means "4 pages", etc.
838 Zero disables swap readahead completely. 736 Zero disables swap readahead completely.
839 737
840 The default value is three (eight pages at a t 738 The default value is three (eight pages at a time). There may be some
841 small benefits in tuning this to a different v 739 small benefits in tuning this to a different value if your workload is
842 swap-intensive. 740 swap-intensive.
843 741
844 Lower values mean lower latencies for initial 742 Lower values mean lower latencies for initial faults, but at the same time
845 extra faults and I/O delays for following faul 743 extra faults and I/O delays for following faults if they would have been part of
846 that consecutive pages readahead would have br 744 that consecutive pages readahead would have brought in.
847 745
848 746
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 747 panic_on_oom
858 ============ 748 ============
859 749
860 This enables or disables panic on out-of-memor 750 This enables or disables panic on out-of-memory feature.
861 751
862 If this is set to 0, the kernel will kill some 752 If this is set to 0, the kernel will kill some rogue process,
863 called oom_killer. Usually, oom_killer can ki 753 called oom_killer. Usually, oom_killer can kill rogue processes and
864 system will survive. 754 system will survive.
865 755
866 If this is set to 1, the kernel panics when ou 756 If this is set to 1, the kernel panics when out-of-memory happens.
867 However, if a process limits using nodes by me 757 However, if a process limits using nodes by mempolicy/cpusets,
868 and those nodes become memory exhaustion statu 758 and those nodes become memory exhaustion status, one process
869 may be killed by oom-killer. No panic occurs i 759 may be killed by oom-killer. No panic occurs in this case.
870 Because other nodes' memory may be free. This 760 Because other nodes' memory may be free. This means system total status
871 may be not fatal yet. 761 may be not fatal yet.
872 762
873 If this is set to 2, the kernel panics compuls 763 If this is set to 2, the kernel panics compulsorily even on the
874 above-mentioned. Even oom happens under memory 764 above-mentioned. Even oom happens under memory cgroup, the whole
875 system panics. 765 system panics.
876 766
877 The default value is 0. 767 The default value is 0.
878 768
879 1 and 2 are for failover of clustering. Please 769 1 and 2 are for failover of clustering. Please select either
880 according to your policy of failover. 770 according to your policy of failover.
881 771
882 panic_on_oom=2+kdump gives you very strong too 772 panic_on_oom=2+kdump gives you very strong tool to investigate
883 why oom happens. You can get snapshot. 773 why oom happens. You can get snapshot.
884 774
885 775
886 percpu_pagelist_high_fraction !! 776 percpu_pagelist_fraction
887 ============================= !! 777 ========================
888 778
889 This is the fraction of pages in each zone tha !! 779 This is the fraction of pages at most (high mark pcp->high) in each zone that
890 per-cpu page lists. It is an upper boundary th !! 780 are allocated for each per cpu page list. The min value for this is 8. It
891 on the number of online CPUs. The min value fo !! 781 means that we don't allow more than 1/8th of pages in each zone to be
892 that we do not allow more than 1/8th of pages !! 782 allocated in any single per_cpu_pagelist. This entry only changes the value
893 on per-cpu page lists. This entry only changes !! 783 of hot per cpu pagelists. User can specify a number like 100 to allocate
894 page lists. A user can specify a number like 1 !! 784 1/100th of each zone to each per cpu page list.
895 each zone between per-cpu lists. !! 785
896 !! 786 The batch value of each per cpu pagelist is also updated as a result. It is
897 The batch value of each per-cpu page list rema !! 787 set to pcp->high/4. The upper limit of batch is (PAGE_SHIFT * 8)
898 the value of the high fraction so allocation l !! 788
899 !! 789 The initial value is zero. Kernel does not use this value at boot time to set
900 The initial value is zero. Kernel uses this va !! 790 the high water marks for each per cpu page list. If the user writes '0' to this
901 mark based on the low watermark for the zone a !! 791 sysctl, it will revert to this default behavior.
902 online CPUs. If the user writes '0' to this s <<
903 this default behavior. <<
904 792
905 793
906 stat_interval 794 stat_interval
907 ============= 795 =============
908 796
909 The time interval between which vm statistics 797 The time interval between which vm statistics are updated. The default
910 is 1 second. 798 is 1 second.
911 799
912 800
913 stat_refresh 801 stat_refresh
914 ============ 802 ============
915 803
916 Any read or write (by root only) flushes all t 804 Any read or write (by root only) flushes all the per-cpu vm statistics
917 into their global totals, for more accurate re 805 into their global totals, for more accurate reports when testing
918 e.g. cat /proc/sys/vm/stat_refresh /proc/memin 806 e.g. cat /proc/sys/vm/stat_refresh /proc/meminfo
919 807
920 As a side-effect, it also checks for negative 808 As a side-effect, it also checks for negative totals (elsewhere reported
921 as 0) and "fails" with EINVAL if any are found 809 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 810 (At time of writing, a few stats are known sometimes to be found negative,
923 with no ill effects: errors and warnings on th 811 with no ill effects: errors and warnings on these stats are suppressed.)
924 812
925 813
926 numa_stat 814 numa_stat
927 ========= 815 =========
928 816
929 This interface allows runtime configuration of 817 This interface allows runtime configuration of numa statistics.
930 818
931 When page allocation performance becomes a bot 819 When page allocation performance becomes a bottleneck and you can tolerate
932 some possible tool breakage and decreased numa 820 some possible tool breakage and decreased numa counter precision, you can
933 do:: 821 do::
934 822
935 echo 0 > /proc/sys/vm/numa_stat 823 echo 0 > /proc/sys/vm/numa_stat
936 824
937 When page allocation performance is not a bott 825 When page allocation performance is not a bottleneck and you want all
938 tooling to work, you can do:: 826 tooling to work, you can do::
939 827
940 echo 1 > /proc/sys/vm/numa_stat 828 echo 1 > /proc/sys/vm/numa_stat
941 829
942 830
943 swappiness 831 swappiness
944 ========== 832 ==========
945 833
946 This control is used to define the rough relat 834 This control is used to define the rough relative IO cost of swapping
947 and filesystem paging, as a value between 0 an 835 and filesystem paging, as a value between 0 and 200. At 100, the VM
948 assumes equal IO cost and will thus apply memo 836 assumes equal IO cost and will thus apply memory pressure to the page
949 cache and swap-backed pages equally; lower val 837 cache and swap-backed pages equally; lower values signify more
950 expensive swap IO, higher values indicates che 838 expensive swap IO, higher values indicates cheaper.
951 839
952 Keep in mind that filesystem IO patterns under 840 Keep in mind that filesystem IO patterns under memory pressure tend to
953 be more efficient than swap's random IO. An op 841 be more efficient than swap's random IO. An optimal value will require
954 experimentation and will also be workload-depe 842 experimentation and will also be workload-dependent.
955 843
956 The default value is 60. 844 The default value is 60.
957 845
958 For in-memory swap, like zram or zswap, as wel 846 For in-memory swap, like zram or zswap, as well as hybrid setups that
959 have swap on faster devices than the filesyste 847 have swap on faster devices than the filesystem, values beyond 100 can
960 be considered. For example, if the random IO a 848 be considered. For example, if the random IO against the swap device
961 is on average 2x faster than IO from the files 849 is on average 2x faster than IO from the filesystem, swappiness should
962 be 133 (x + 2x = 200, 2x = 133.33). 850 be 133 (x + 2x = 200, 2x = 133.33).
963 851
964 At 0, the kernel will not initiate swap until 852 At 0, the kernel will not initiate swap until the amount of free and
965 file-backed pages is less than the high waterm 853 file-backed pages is less than the high watermark in a zone.
966 854
967 855
968 unprivileged_userfaultfd 856 unprivileged_userfaultfd
969 ======================== 857 ========================
970 858
971 This flag controls the mode in which unprivile !! 859 This flag controls whether unprivileged users can use the userfaultfd
972 userfaultfd system calls. Set this to 0 to res !! 860 system calls. Set this to 1 to allow unprivileged users to use the
973 to handle page faults in user mode only. In th !! 861 userfaultfd system calls, or set this to 0 to restrict userfaultfd to only
974 SYS_CAP_PTRACE must pass UFFD_USER_MODE_ONLY i !! 862 privileged users (with SYS_CAP_PTRACE capability).
975 succeed. Prohibiting use of userfaultfd for ha <<
976 mode may make certain vulnerabilities more dif <<
977 <<
978 Set this to 1 to allow unprivileged users to u <<
979 calls without any restrictions. <<
980 863
981 The default value is 0. !! 864 The default value is 1.
982 865
983 Another way to control permissions for userfau <<
984 /dev/userfaultfd instead of userfaultfd(2). Se <<
985 Documentation/admin-guide/mm/userfaultfd.rst. <<
986 866
987 user_reserve_kbytes 867 user_reserve_kbytes
988 =================== 868 ===================
989 869
990 When overcommit_memory is set to 2, "never ove 870 When overcommit_memory is set to 2, "never overcommit" mode, reserve
991 min(3% of current process size, user_reserve_k 871 min(3% of current process size, user_reserve_kbytes) of free memory.
992 This is intended to prevent a user from starti 872 This is intended to prevent a user from starting a single memory hogging
993 process, such that they cannot recover (kill t 873 process, such that they cannot recover (kill the hog).
994 874
995 user_reserve_kbytes defaults to min(3% of the 875 user_reserve_kbytes defaults to min(3% of the current process size, 128MB).
996 876
997 If this is reduced to zero, then the user will 877 If this is reduced to zero, then the user will be allowed to allocate
998 all free memory with a single process, minus a 878 all free memory with a single process, minus admin_reserve_kbytes.
999 Any subsequent attempts to execute a command w 879 Any subsequent attempts to execute a command will result in
1000 "fork: Cannot allocate memory". 880 "fork: Cannot allocate memory".
1001 881
1002 Changing this takes effect whenever an applic 882 Changing this takes effect whenever an application requests memory.
1003 883
1004 884
1005 vfs_cache_pressure 885 vfs_cache_pressure
1006 ================== 886 ==================
1007 887
1008 This percentage value controls the tendency o 888 This percentage value controls the tendency of the kernel to reclaim
1009 the memory which is used for caching of direc 889 the memory which is used for caching of directory and inode objects.
1010 890
1011 At the default value of vfs_cache_pressure=10 891 At the default value of vfs_cache_pressure=100 the kernel will attempt to
1012 reclaim dentries and inodes at a "fair" rate 892 reclaim dentries and inodes at a "fair" rate with respect to pagecache and
1013 swapcache reclaim. Decreasing vfs_cache_pres 893 swapcache reclaim. Decreasing vfs_cache_pressure causes the kernel to prefer
1014 to retain dentry and inode caches. When vfs_c 894 to retain dentry and inode caches. When vfs_cache_pressure=0, the kernel will
1015 never reclaim dentries and inodes due to memo 895 never reclaim dentries and inodes due to memory pressure and this can easily
1016 lead to out-of-memory conditions. Increasing 896 lead to out-of-memory conditions. Increasing vfs_cache_pressure beyond 100
1017 causes the kernel to prefer to reclaim dentri 897 causes the kernel to prefer to reclaim dentries and inodes.
1018 898
1019 Increasing vfs_cache_pressure significantly b 899 Increasing vfs_cache_pressure significantly beyond 100 may have negative
1020 performance impact. Reclaim code needs to tak 900 performance impact. Reclaim code needs to take various locks to find freeable
1021 directory and inode objects. With vfs_cache_p 901 directory and inode objects. With vfs_cache_pressure=1000, it will look for
1022 ten times more freeable objects than there ar 902 ten times more freeable objects than there are.
1023 903
1024 904
1025 watermark_boost_factor 905 watermark_boost_factor
1026 ====================== 906 ======================
1027 907
1028 This factor controls the level of reclaim whe 908 This factor controls the level of reclaim when memory is being fragmented.
1029 It defines the percentage of the high waterma 909 It defines the percentage of the high watermark of a zone that will be
1030 reclaimed if pages of different mobility are 910 reclaimed if pages of different mobility are being mixed within pageblocks.
1031 The intent is that compaction has less work t 911 The intent is that compaction has less work to do in the future and to
1032 increase the success rate of future high-orde 912 increase the success rate of future high-order allocations such as SLUB
1033 allocations, THP and hugetlbfs pages. 913 allocations, THP and hugetlbfs pages.
1034 914
1035 To make it sensible with respect to the water 915 To make it sensible with respect to the watermark_scale_factor
1036 parameter, the unit is in fractions of 10,000 916 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 !! 917 15,000 on !DISCONTIGMEM configurations means that up to 150% of the high
1038 event of a pageblock being mixed due to fragm !! 918 watermark will be reclaimed in the event of a pageblock being mixed due
1039 is determined by the number of fragmentation !! 919 to fragmentation. The level of reclaim is determined by the number of
1040 recent past. If this value is smaller than a !! 920 fragmentation events that occurred in the recent past. If this value is
1041 worth of pages will be reclaimed (e.g. 2MB o !! 921 smaller than a pageblock then a pageblocks worth of pages will be reclaimed
1042 of 0 will disable the feature. !! 922 (e.g. 2MB on 64-bit x86). A boost factor of 0 will disable the feature.
1043 923
1044 924
1045 watermark_scale_factor 925 watermark_scale_factor
1046 ====================== 926 ======================
1047 927
1048 This factor controls the aggressiveness of ks 928 This factor controls the aggressiveness of kswapd. It defines the
1049 amount of memory left in a node/system before 929 amount of memory left in a node/system before kswapd is woken up and
1050 how much memory needs to be free before kswap 930 how much memory needs to be free before kswapd goes back to sleep.
1051 931
1052 The unit is in fractions of 10,000. The defau 932 The unit is in fractions of 10,000. The default value of 10 means the
1053 distances between watermarks are 0.1% of the 933 distances between watermarks are 0.1% of the available memory in the
1054 node/system. The maximum value is 3000, or 30 !! 934 node/system. The maximum value is 1000, or 10% of memory.
1055 935
1056 A high rate of threads entering direct reclai 936 A high rate of threads entering direct reclaim (allocstall) or kswapd
1057 going to sleep prematurely (kswapd_low_wmark_ 937 going to sleep prematurely (kswapd_low_wmark_hit_quickly) can indicate
1058 that the number of free pages kswapd maintain 938 that the number of free pages kswapd maintains for latency reasons is
1059 too small for the allocation bursts occurring 939 too small for the allocation bursts occurring in the system. This knob
1060 can then be used to tune kswapd aggressivenes 940 can then be used to tune kswapd aggressiveness accordingly.
1061 941
1062 942
1063 zone_reclaim_mode 943 zone_reclaim_mode
1064 ================= 944 =================
1065 945
1066 Zone_reclaim_mode allows someone to set more 946 Zone_reclaim_mode allows someone to set more or less aggressive approaches to
1067 reclaim memory when a zone runs out of memory 947 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 948 zone reclaim occurs. Allocations will be satisfied from other zones / nodes
1069 in the system. 949 in the system.
1070 950
1071 This is value OR'ed together of 951 This is value OR'ed together of
1072 952
1073 = =================================== 953 = ===================================
1074 1 Zone reclaim on 954 1 Zone reclaim on
1075 2 Zone reclaim writes dirty pages out 955 2 Zone reclaim writes dirty pages out
1076 4 Zone reclaim swaps pages 956 4 Zone reclaim swaps pages
1077 = =================================== 957 = ===================================
1078 958
1079 zone_reclaim_mode is disabled by default. Fo 959 zone_reclaim_mode is disabled by default. For file servers or workloads
1080 that benefit from having their data cached, z 960 that benefit from having their data cached, zone_reclaim_mode should be
1081 left disabled as the caching effect is likely 961 left disabled as the caching effect is likely to be more important than
1082 data locality. 962 data locality.
1083 963
1084 Consider enabling one or more zone_reclaim mo !! 964 zone_reclaim may be enabled if it's known that the workload is partitioned
1085 workload is partitioned such that each partit !! 965 such that each partition fits within a NUMA node and that accessing remote
1086 and that accessing remote memory would cause !! 966 memory would cause a measurable performance reduction. The page allocator
1087 reduction. The page allocator will take addi !! 967 will then reclaim easily reusable pages (those page cache pages that are
1088 allocating off node pages. !! 968 currently not used) before allocating off node pages.
1089 969
1090 Allowing zone reclaim to write out pages stop 970 Allowing zone reclaim to write out pages stops processes that are
1091 writing large amounts of data from dirtying p 971 writing large amounts of data from dirtying pages on other nodes. Zone
1092 reclaim will write out dirty pages if a zone 972 reclaim will write out dirty pages if a zone fills up and so effectively
1093 throttle the process. This may decrease the p 973 throttle the process. This may decrease the performance of a single process
1094 since it cannot use all of system memory to b 974 since it cannot use all of system memory to buffer the outgoing writes
1095 anymore but it preserve the memory on other n 975 anymore but it preserve the memory on other nodes so that the performance
1096 of other processes running on other nodes wil 976 of other processes running on other nodes will not be affected.
1097 977
1098 Allowing regular swap effectively restricts a 978 Allowing regular swap effectively restricts allocations to the local
1099 node unless explicitly overridden by memory p 979 node unless explicitly overridden by memory policies or cpuset
1100 configurations. 980 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.