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