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