1 ======================================== 1 ========================================
2 zram: Compressed RAM-based block devices 2 zram: Compressed RAM-based block devices
3 ======================================== 3 ========================================
4 4
5 Introduction 5 Introduction
6 ============ 6 ============
7 7
8 The zram module creates RAM-based block device 8 The zram module creates RAM-based block devices named /dev/zram<id>
9 (<id> = 0, 1, ...). Pages written to these dis 9 (<id> = 0, 1, ...). Pages written to these disks are compressed and stored
10 in memory itself. These disks allow very fast 10 in memory itself. These disks allow very fast I/O and compression provides
11 good amounts of memory savings. Some of the us 11 good amounts of memory savings. Some of the use cases include /tmp storage,
12 use as swap disks, various caches under /var a 12 use as swap disks, various caches under /var and maybe many more. :)
13 13
14 Statistics for individual zram devices are exp 14 Statistics for individual zram devices are exported through sysfs nodes at
15 /sys/block/zram<id>/ 15 /sys/block/zram<id>/
16 16
17 Usage 17 Usage
18 ===== 18 =====
19 19
20 There are several ways to configure and manage 20 There are several ways to configure and manage zram device(-s):
21 21
22 a) using zram and zram_control sysfs attribute 22 a) using zram and zram_control sysfs attributes
23 b) using zramctl utility, provided by util-lin 23 b) using zramctl utility, provided by util-linux (util-linux@vger.kernel.org).
24 24
25 In this document we will describe only 'manual 25 In this document we will describe only 'manual' zram configuration steps,
26 IOW, zram and zram_control sysfs attributes. 26 IOW, zram and zram_control sysfs attributes.
27 27
28 In order to get a better idea about zramctl pl 28 In order to get a better idea about zramctl please consult util-linux
29 documentation, zramctl man-page or `zramctl -- 29 documentation, zramctl man-page or `zramctl --help`. Please be informed
30 that zram maintainers do not develop/maintain 30 that zram maintainers do not develop/maintain util-linux or zramctl, should
31 you have any questions please contact util-lin 31 you have any questions please contact util-linux@vger.kernel.org
32 32
33 Following shows a typical sequence of steps fo 33 Following shows a typical sequence of steps for using zram.
34 34
35 WARNING 35 WARNING
36 ======= 36 =======
37 37
38 For the sake of simplicity we skip error check 38 For the sake of simplicity we skip error checking parts in most of the
39 examples below. However, it is your sole respo 39 examples below. However, it is your sole responsibility to handle errors.
40 40
41 zram sysfs attributes always return negative v 41 zram sysfs attributes always return negative values in case of errors.
42 The list of possible return codes: 42 The list of possible return codes:
43 43
44 ======== ==================================== 44 ======== =============================================================
45 -EBUSY an attempt to modify an attribute th 45 -EBUSY an attempt to modify an attribute that cannot be changed once
46 the device has been initialised. Ple 46 the device has been initialised. Please reset device first.
47 -ENOMEM zram was not able to allocate enough 47 -ENOMEM zram was not able to allocate enough memory to fulfil your
48 needs. 48 needs.
49 -EINVAL invalid input has been provided. 49 -EINVAL invalid input has been provided.
50 ======== ==================================== 50 ======== =============================================================
51 51
52 If you use 'echo', the returned value is set b 52 If you use 'echo', the returned value is set by the 'echo' utility,
53 and, in general case, something like:: 53 and, in general case, something like::
54 54
55 echo 3 > /sys/block/zram0/max_comp_str 55 echo 3 > /sys/block/zram0/max_comp_streams
56 if [ $? -ne 0 ]; then 56 if [ $? -ne 0 ]; then
57 handle_error 57 handle_error
58 fi 58 fi
59 59
60 should suffice. 60 should suffice.
61 61
62 1) Load Module 62 1) Load Module
63 ============== 63 ==============
64 64
65 :: 65 ::
66 66
67 modprobe zram num_devices=4 67 modprobe zram num_devices=4
68 68
69 This creates 4 devices: /dev/zram{0,1,2,3} 69 This creates 4 devices: /dev/zram{0,1,2,3}
70 70
71 num_devices parameter is optional and tells zr 71 num_devices parameter is optional and tells zram how many devices should be
72 pre-created. Default: 1. 72 pre-created. Default: 1.
73 73
74 2) Set max number of compression streams 74 2) Set max number of compression streams
75 ======================================== 75 ========================================
76 76
77 Regardless of the value passed to this attribu 77 Regardless of the value passed to this attribute, ZRAM will always
78 allocate multiple compression streams - one pe 78 allocate multiple compression streams - one per online CPU - thus
79 allowing several concurrent compression operat 79 allowing several concurrent compression operations. The number of
80 allocated compression streams goes down when s 80 allocated compression streams goes down when some of the CPUs
81 become offline. There is no single-compression 81 become offline. There is no single-compression-stream mode anymore,
82 unless you are running a UP system or have onl 82 unless you are running a UP system or have only 1 CPU online.
83 83
84 To find out how many streams are currently ava 84 To find out how many streams are currently available::
85 85
86 cat /sys/block/zram0/max_comp_streams 86 cat /sys/block/zram0/max_comp_streams
87 87
88 3) Select compression algorithm 88 3) Select compression algorithm
89 =============================== 89 ===============================
90 90
91 Using comp_algorithm device attribute one can 91 Using comp_algorithm device attribute one can see available and
92 currently selected (shown in square brackets) 92 currently selected (shown in square brackets) compression algorithms,
93 or change the selected compression algorithm ( 93 or change the selected compression algorithm (once the device is initialised
94 there is no way to change compression algorith 94 there is no way to change compression algorithm).
95 95
96 Examples:: 96 Examples::
97 97
98 #show supported compression algorithms 98 #show supported compression algorithms
99 cat /sys/block/zram0/comp_algorithm 99 cat /sys/block/zram0/comp_algorithm
100 lzo [lz4] 100 lzo [lz4]
101 101
102 #select lzo compression algorithm 102 #select lzo compression algorithm
103 echo lzo > /sys/block/zram0/comp_algor 103 echo lzo > /sys/block/zram0/comp_algorithm
104 104
105 For the time being, the `comp_algorithm` conte !! 105 For the time being, the `comp_algorithm` content does not necessarily
106 algorithms that are supported by zram. !! 106 show every compression algorithm supported by the kernel. We keep this
>> 107 list primarily to simplify device configuration and one can configure
>> 108 a new device with a compression algorithm that is not listed in
>> 109 `comp_algorithm`. The thing is that, internally, ZRAM uses Crypto API
>> 110 and, if some of the algorithms were built as modules, it's impossible
>> 111 to list all of them using, for instance, /proc/crypto or any other
>> 112 method. This, however, has an advantage of permitting the usage of
>> 113 custom crypto compression modules (implementing S/W or H/W compression).
107 114
108 4) Set compression algorithm parameters: Optio !! 115 4) Set Disksize
109 ============================================== <<
110 <<
111 Compression algorithms may support specific pa <<
112 tweaked for particular dataset. ZRAM has an `a <<
113 attribute which provides a per-algorithm param <<
114 <<
115 For example, several compression algorithms su <<
116 In addition, certain compression algorithms su <<
117 which significantly change algorithms' charact <<
118 compression algorithm to use external pre-trai <<
119 path to the `dict` along with other parameters <<
120 <<
121 #pass path to pre-trained zstd diction <<
122 echo "algo=zstd dict=/etc/dictioary" > <<
123 <<
124 #same, but using algorithm priority <<
125 echo "priority=1 dict=/etc/dictioary" <<
126 /sys/block/zram0/algorithm_par <<
127 <<
128 #pass path to pre-trained zstd diction <<
129 echo "algo=zstd level=8 dict=/etc/dict <<
130 /sys/block/zram0/algorithm_par <<
131 <<
132 Parameters are algorithm specific: not all alg <<
133 dictionaries, not all algorithms support `leve <<
134 algorithms `level` controls the compression le <<
135 better the compression ratio, it even can take <<
136 algorithms), for other algorithms `level` is a <<
137 the value the lower the compression ratio). <<
138 <<
139 5) Set Disksize <<
140 =============== 116 ===============
141 117
142 Set disk size by writing the value to sysfs no 118 Set disk size by writing the value to sysfs node 'disksize'.
143 The value can be either in bytes or you can us 119 The value can be either in bytes or you can use mem suffixes.
144 Examples:: 120 Examples::
145 121
146 # Initialize /dev/zram0 with 50MB disk 122 # Initialize /dev/zram0 with 50MB disksize
147 echo $((50*1024*1024)) > /sys/block/zr 123 echo $((50*1024*1024)) > /sys/block/zram0/disksize
148 124
149 # Using mem suffixes 125 # Using mem suffixes
150 echo 256K > /sys/block/zram0/disksize 126 echo 256K > /sys/block/zram0/disksize
151 echo 512M > /sys/block/zram0/disksize 127 echo 512M > /sys/block/zram0/disksize
152 echo 1G > /sys/block/zram0/disksize 128 echo 1G > /sys/block/zram0/disksize
153 129
154 Note: 130 Note:
155 There is little point creating a zram of great 131 There is little point creating a zram of greater than twice the size of memory
156 since we expect a 2:1 compression ratio. Note 132 since we expect a 2:1 compression ratio. Note that zram uses about 0.1% of the
157 size of the disk when not in use so a huge zra 133 size of the disk when not in use so a huge zram is wasteful.
158 134
159 6) Set memory limit: Optional !! 135 5) Set memory limit: Optional
160 ============================= 136 =============================
161 137
162 Set memory limit by writing the value to sysfs 138 Set memory limit by writing the value to sysfs node 'mem_limit'.
163 The value can be either in bytes or you can us 139 The value can be either in bytes or you can use mem suffixes.
164 In addition, you could change the value in run 140 In addition, you could change the value in runtime.
165 Examples:: 141 Examples::
166 142
167 # limit /dev/zram0 with 50MB memory 143 # limit /dev/zram0 with 50MB memory
168 echo $((50*1024*1024)) > /sys/block/zr 144 echo $((50*1024*1024)) > /sys/block/zram0/mem_limit
169 145
170 # Using mem suffixes 146 # Using mem suffixes
171 echo 256K > /sys/block/zram0/mem_limit 147 echo 256K > /sys/block/zram0/mem_limit
172 echo 512M > /sys/block/zram0/mem_limit 148 echo 512M > /sys/block/zram0/mem_limit
173 echo 1G > /sys/block/zram0/mem_limit 149 echo 1G > /sys/block/zram0/mem_limit
174 150
175 # To disable memory limit 151 # To disable memory limit
176 echo 0 > /sys/block/zram0/mem_limit 152 echo 0 > /sys/block/zram0/mem_limit
177 153
178 7) Activate !! 154 6) Activate
179 =========== 155 ===========
180 156
181 :: 157 ::
182 158
183 mkswap /dev/zram0 159 mkswap /dev/zram0
184 swapon /dev/zram0 160 swapon /dev/zram0
185 161
186 mkfs.ext4 /dev/zram1 162 mkfs.ext4 /dev/zram1
187 mount /dev/zram1 /tmp 163 mount /dev/zram1 /tmp
188 164
189 8) Add/remove zram devices !! 165 7) Add/remove zram devices
190 ========================== 166 ==========================
191 167
192 zram provides a control interface, which enabl 168 zram provides a control interface, which enables dynamic (on-demand) device
193 addition and removal. 169 addition and removal.
194 170
195 In order to add a new /dev/zramX device, perfo 171 In order to add a new /dev/zramX device, perform a read operation on the hot_add
196 attribute. This will return either the new dev 172 attribute. This will return either the new device's device id (meaning that you
197 can use /dev/zram<id>) or an error code. 173 can use /dev/zram<id>) or an error code.
198 174
199 Example:: 175 Example::
200 176
201 cat /sys/class/zram-control/hot_add 177 cat /sys/class/zram-control/hot_add
202 1 178 1
203 179
204 To remove the existing /dev/zramX device (wher 180 To remove the existing /dev/zramX device (where X is a device id)
205 execute:: 181 execute::
206 182
207 echo X > /sys/class/zram-control/hot_r 183 echo X > /sys/class/zram-control/hot_remove
208 184
209 9) Stats !! 185 8) Stats
210 ======== 186 ========
211 187
212 Per-device statistics are exported as various 188 Per-device statistics are exported as various nodes under /sys/block/zram<id>/
213 189
214 A brief description of exported device attribu 190 A brief description of exported device attributes follows. For more details
215 please read Documentation/ABI/testing/sysfs-bl 191 please read Documentation/ABI/testing/sysfs-block-zram.
216 192
217 ====================== ====== ============== 193 ====================== ====== ===============================================
218 Name access desc 194 Name access description
219 ====================== ====== ============== 195 ====================== ====== ===============================================
220 disksize RW show and set t 196 disksize RW show and set the device's disk size
221 initstate RO shows the init 197 initstate RO shows the initialization state of the device
222 reset WO trigger device 198 reset WO trigger device reset
223 mem_used_max WO reset the `mem 199 mem_used_max WO reset the `mem_used_max` counter (see later)
224 mem_limit WO specifies the 200 mem_limit WO specifies the maximum amount of memory ZRAM can
225 use to store t 201 use to store the compressed data
226 writeback_limit WO specifies the 202 writeback_limit WO specifies the maximum amount of write IO zram
227 can write out 203 can write out to backing device as 4KB unit
228 writeback_limit_enable RW show and set w 204 writeback_limit_enable RW show and set writeback_limit feature
229 max_comp_streams RW the number of 205 max_comp_streams RW the number of possible concurrent compress
230 operations 206 operations
231 comp_algorithm RW show and chang 207 comp_algorithm RW show and change the compression algorithm
232 algorithm_params WO setup compress <<
233 compact WO trigger memory 208 compact WO trigger memory compaction
234 debug_stat RO this file is u 209 debug_stat RO this file is used for zram debugging purposes
235 backing_dev RW set up backend 210 backing_dev RW set up backend storage for zram to write out
236 idle WO mark allocated 211 idle WO mark allocated slot as idle
237 ====================== ====== ============== 212 ====================== ====== ===============================================
238 213
239 214
240 User space is advised to use the following fil 215 User space is advised to use the following files to read the device statistics.
241 216
242 File /sys/block/zram<id>/stat 217 File /sys/block/zram<id>/stat
243 218
244 Represents block layer statistics. Read Docume 219 Represents block layer statistics. Read Documentation/block/stat.rst for
245 details. 220 details.
246 221
247 File /sys/block/zram<id>/io_stat 222 File /sys/block/zram<id>/io_stat
248 223
249 The stat file represents device's I/O statisti 224 The stat file represents device's I/O statistics not accounted by block
250 layer and, thus, not available in zram<id>/sta 225 layer and, thus, not available in zram<id>/stat file. It consists of a
251 single line of text and contains the following 226 single line of text and contains the following stats separated by
252 whitespace: 227 whitespace:
253 228
254 ============= ============================ 229 ============= =============================================================
255 failed_reads The number of failed reads 230 failed_reads The number of failed reads
256 failed_writes The number of failed writes 231 failed_writes The number of failed writes
257 invalid_io The number of non-page-size- 232 invalid_io The number of non-page-size-aligned I/O requests
258 notify_free Depending on device usage sc 233 notify_free Depending on device usage scenario it may account
259 234
260 a) the number of pages freed 235 a) the number of pages freed because of swap slot free
261 notifications 236 notifications
262 b) the number of pages freed 237 b) the number of pages freed because of
263 REQ_OP_DISCARD requests s 238 REQ_OP_DISCARD requests sent by bio. The former ones are
264 sent to a swap block devi 239 sent to a swap block device when a swap slot is freed,
265 which implies that this d 240 which implies that this disk is being used as a swap disk.
266 241
267 The latter ones are sent by 242 The latter ones are sent by filesystem mounted with
268 discard option, whenever som 243 discard option, whenever some data blocks are getting
269 discarded. 244 discarded.
270 ============= ============================ 245 ============= =============================================================
271 246
272 File /sys/block/zram<id>/mm_stat 247 File /sys/block/zram<id>/mm_stat
273 248
274 The mm_stat file represents the device's mm st 249 The mm_stat file represents the device's mm statistics. It consists of a single
275 line of text and contains the following stats 250 line of text and contains the following stats separated by whitespace:
276 251
277 ================ ============================ 252 ================ =============================================================
278 orig_data_size uncompressed size of data st 253 orig_data_size uncompressed size of data stored in this disk.
279 Unit: bytes 254 Unit: bytes
280 compr_data_size compressed size of data stor 255 compr_data_size compressed size of data stored in this disk
281 mem_used_total the amount of memory allocat 256 mem_used_total the amount of memory allocated for this disk. This
282 includes allocator fragmenta 257 includes allocator fragmentation and metadata overhead,
283 allocated for this disk. So, 258 allocated for this disk. So, allocator space efficiency
284 can be calculated using comp 259 can be calculated using compr_data_size and this statistic.
285 Unit: bytes 260 Unit: bytes
286 mem_limit the maximum amount of memory 261 mem_limit the maximum amount of memory ZRAM can use to store
287 the compressed data 262 the compressed data
288 mem_used_max the maximum amount of memory 263 mem_used_max the maximum amount of memory zram has consumed to
289 store the data 264 store the data
290 same_pages the number of same element f 265 same_pages the number of same element filled pages written to this disk.
291 No memory is allocated for s 266 No memory is allocated for such pages.
292 pages_compacted the number of pages freed du 267 pages_compacted the number of pages freed during compaction
293 huge_pages the number of incompressible 268 huge_pages the number of incompressible pages
294 huge_pages_since the number of incompressible 269 huge_pages_since the number of incompressible pages since zram set up
295 ================ ============================ 270 ================ =============================================================
296 271
297 File /sys/block/zram<id>/bd_stat 272 File /sys/block/zram<id>/bd_stat
298 273
299 The bd_stat file represents a device's backing 274 The bd_stat file represents a device's backing device statistics. It consists of
300 a single line of text and contains the followi 275 a single line of text and contains the following stats separated by whitespace:
301 276
302 ============== ============================== 277 ============== =============================================================
303 bd_count size of data written in backin 278 bd_count size of data written in backing device.
304 Unit: 4K bytes 279 Unit: 4K bytes
305 bd_reads the number of reads from backi 280 bd_reads the number of reads from backing device
306 Unit: 4K bytes 281 Unit: 4K bytes
307 bd_writes the number of writes to backin 282 bd_writes the number of writes to backing device
308 Unit: 4K bytes 283 Unit: 4K bytes
309 ============== ============================== 284 ============== =============================================================
310 285
311 10) Deactivate !! 286 9) Deactivate
312 ============== !! 287 =============
313 288
314 :: 289 ::
315 290
316 swapoff /dev/zram0 291 swapoff /dev/zram0
317 umount /dev/zram1 292 umount /dev/zram1
318 293
319 11) Reset !! 294 10) Reset
320 ========= 295 =========
321 296
322 Write any positive value to 'reset' sy 297 Write any positive value to 'reset' sysfs node::
323 298
324 echo 1 > /sys/block/zram0/rese 299 echo 1 > /sys/block/zram0/reset
325 echo 1 > /sys/block/zram1/rese 300 echo 1 > /sys/block/zram1/reset
326 301
327 This frees all the memory allocated fo 302 This frees all the memory allocated for the given device and
328 resets the disksize to zero. You must 303 resets the disksize to zero. You must set the disksize again
329 before reusing the device. 304 before reusing the device.
330 305
331 Optional Feature 306 Optional Feature
332 ================ 307 ================
333 308
334 writeback 309 writeback
335 --------- 310 ---------
336 311
337 With CONFIG_ZRAM_WRITEBACK, zram can write idl 312 With CONFIG_ZRAM_WRITEBACK, zram can write idle/incompressible page
338 to backing storage rather than keeping it in m 313 to backing storage rather than keeping it in memory.
339 To use the feature, admin should set up backin 314 To use the feature, admin should set up backing device via::
340 315
341 echo /dev/sda5 > /sys/block/zramX/back 316 echo /dev/sda5 > /sys/block/zramX/backing_dev
342 317
343 before disksize setting. It supports only part 318 before disksize setting. It supports only partitions at this moment.
344 If admin wants to use incompressible page writ 319 If admin wants to use incompressible page writeback, they could do it via::
345 320
346 echo huge > /sys/block/zramX/writeback 321 echo huge > /sys/block/zramX/writeback
347 322
348 To use idle page writeback, first, user need t 323 To use idle page writeback, first, user need to declare zram pages
349 as idle:: 324 as idle::
350 325
351 echo all > /sys/block/zramX/idle 326 echo all > /sys/block/zramX/idle
352 327
353 From now on, any pages on zram are idle pages. 328 From now on, any pages on zram are idle pages. The idle mark
354 will be removed until someone requests access 329 will be removed until someone requests access of the block.
355 IOW, unless there is access request, those pag 330 IOW, unless there is access request, those pages are still idle pages.
356 Additionally, when CONFIG_ZRAM_TRACK_ENTRY_ACT !! 331 Additionally, when CONFIG_ZRAM_MEMORY_TRACKING is enabled pages can be
357 marked as idle based on how long (in seconds) 332 marked as idle based on how long (in seconds) it's been since they were
358 last accessed:: 333 last accessed::
359 334
360 echo 86400 > /sys/block/zramX/idle 335 echo 86400 > /sys/block/zramX/idle
361 336
362 In this example all pages which haven't been a 337 In this example all pages which haven't been accessed in more than 86400
363 seconds (one day) will be marked idle. 338 seconds (one day) will be marked idle.
364 339
365 Admin can request writeback of those idle page 340 Admin can request writeback of those idle pages at right timing via::
366 341
367 echo idle > /sys/block/zramX/writeback 342 echo idle > /sys/block/zramX/writeback
368 343
369 With the command, zram will writeback idle pag 344 With the command, zram will writeback idle pages from memory to the storage.
370 345
371 Additionally, if a user choose to writeback on 346 Additionally, if a user choose to writeback only huge and idle pages
372 this can be accomplished with:: 347 this can be accomplished with::
373 348
374 echo huge_idle > /sys/block/zramX/writ 349 echo huge_idle > /sys/block/zramX/writeback
375 350
376 If a user chooses to writeback only incompress 351 If a user chooses to writeback only incompressible pages (pages that none of
377 algorithms can compress) this can be accomplis 352 algorithms can compress) this can be accomplished with::
378 353
379 echo incompressible > /sys/block/zramX 354 echo incompressible > /sys/block/zramX/writeback
380 355
381 If an admin wants to write a specific page in 356 If an admin wants to write a specific page in zram device to the backing device,
382 they could write a page index into the interfa 357 they could write a page index into the interface::
383 358
384 echo "page_index=1251" > /sys/block/zr 359 echo "page_index=1251" > /sys/block/zramX/writeback
385 360
386 If there are lots of write IO with flash devic 361 If there are lots of write IO with flash device, potentially, it has
387 flash wearout problem so that admin needs to d 362 flash wearout problem so that admin needs to design write limitation
388 to guarantee storage health for entire product 363 to guarantee storage health for entire product life.
389 364
390 To overcome the concern, zram supports "writeb 365 To overcome the concern, zram supports "writeback_limit" feature.
391 The "writeback_limit_enable"'s default value i 366 The "writeback_limit_enable"'s default value is 0 so that it doesn't limit
392 any writeback. IOW, if admin wants to apply wr 367 any writeback. IOW, if admin wants to apply writeback budget, they should
393 enable writeback_limit_enable via:: 368 enable writeback_limit_enable via::
394 369
395 $ echo 1 > /sys/block/zramX/writeback_ 370 $ echo 1 > /sys/block/zramX/writeback_limit_enable
396 371
397 Once writeback_limit_enable is set, zram doesn 372 Once writeback_limit_enable is set, zram doesn't allow any writeback
398 until admin sets the budget via /sys/block/zra 373 until admin sets the budget via /sys/block/zramX/writeback_limit.
399 374
400 (If admin doesn't enable writeback_limit_enabl 375 (If admin doesn't enable writeback_limit_enable, writeback_limit's value
401 assigned via /sys/block/zramX/writeback_limit 376 assigned via /sys/block/zramX/writeback_limit is meaningless.)
402 377
403 If admin wants to limit writeback as per-day 4 378 If admin wants to limit writeback as per-day 400M, they could do it
404 like below:: 379 like below::
405 380
406 $ MB_SHIFT=20 381 $ MB_SHIFT=20
407 $ 4K_SHIFT=12 382 $ 4K_SHIFT=12
408 $ echo $((400<<MB_SHIFT>>4K_SHIFT)) > 383 $ echo $((400<<MB_SHIFT>>4K_SHIFT)) > \
409 /sys/block/zram0/writeback_lim 384 /sys/block/zram0/writeback_limit.
410 $ echo 1 > /sys/block/zram0/writeback_ 385 $ echo 1 > /sys/block/zram0/writeback_limit_enable
411 386
412 If admins want to allow further write again on 387 If admins want to allow further write again once the budget is exhausted,
413 they could do it like below:: 388 they could do it like below::
414 389
415 $ echo $((400<<MB_SHIFT>>4K_SHIFT)) > 390 $ echo $((400<<MB_SHIFT>>4K_SHIFT)) > \
416 /sys/block/zram0/writeback_lim 391 /sys/block/zram0/writeback_limit
417 392
418 If an admin wants to see the remaining writeba 393 If an admin wants to see the remaining writeback budget since last set::
419 394
420 $ cat /sys/block/zramX/writeback_limit 395 $ cat /sys/block/zramX/writeback_limit
421 396
422 If an admin wants to disable writeback limit, 397 If an admin wants to disable writeback limit, they could do::
423 398
424 $ echo 0 > /sys/block/zramX/writeback_ 399 $ echo 0 > /sys/block/zramX/writeback_limit_enable
425 400
426 The writeback_limit count will reset whenever 401 The writeback_limit count will reset whenever you reset zram (e.g.,
427 system reboot, echo 1 > /sys/block/zramX/reset 402 system reboot, echo 1 > /sys/block/zramX/reset) so keeping how many of
428 writeback happened until you reset the zram to 403 writeback happened until you reset the zram to allocate extra writeback
429 budget in next setting is user's job. 404 budget in next setting is user's job.
430 405
431 If admin wants to measure writeback count in a 406 If admin wants to measure writeback count in a certain period, they could
432 know it via /sys/block/zram0/bd_stat's 3rd col 407 know it via /sys/block/zram0/bd_stat's 3rd column.
433 408
434 recompression 409 recompression
435 ------------- 410 -------------
436 411
437 With CONFIG_ZRAM_MULTI_COMP, zram can recompre 412 With CONFIG_ZRAM_MULTI_COMP, zram can recompress pages using alternative
438 (secondary) compression algorithms. The basic 413 (secondary) compression algorithms. The basic idea is that alternative
439 compression algorithm can provide better compr 414 compression algorithm can provide better compression ratio at a price of
440 (potentially) slower compression/decompression 415 (potentially) slower compression/decompression speeds. Alternative compression
441 algorithm can, for example, be more successful 416 algorithm can, for example, be more successful compressing huge pages (those
442 that default algorithm failed to compress). An 417 that default algorithm failed to compress). Another application is idle pages
443 recompression - pages that are cold and sit in 418 recompression - pages that are cold and sit in the memory can be recompressed
444 using more effective algorithm and, hence, red 419 using more effective algorithm and, hence, reduce zsmalloc memory usage.
445 420
446 With CONFIG_ZRAM_MULTI_COMP, zram supports up 421 With CONFIG_ZRAM_MULTI_COMP, zram supports up to 4 compression algorithms:
447 one primary and up to 3 secondary ones. Primar 422 one primary and up to 3 secondary ones. Primary zram compressor is explained
448 in "3) Select compression algorithm", secondar 423 in "3) Select compression algorithm", secondary algorithms are configured
449 using recomp_algorithm device attribute. 424 using recomp_algorithm device attribute.
450 425
451 Example::: 426 Example:::
452 427
453 #show supported recompression algorith 428 #show supported recompression algorithms
454 cat /sys/block/zramX/recomp_algorithm 429 cat /sys/block/zramX/recomp_algorithm
455 #1: lzo lzo-rle lz4 lz4hc [zstd] 430 #1: lzo lzo-rle lz4 lz4hc [zstd]
456 #2: lzo lzo-rle lz4 [lz4hc] zstd 431 #2: lzo lzo-rle lz4 [lz4hc] zstd
457 432
458 Alternative compression algorithms are sorted 433 Alternative compression algorithms are sorted by priority. In the example
459 above, zstd is used as the first alternative a 434 above, zstd is used as the first alternative algorithm, which has priority
460 of 1, while lz4hc is configured as a compressi 435 of 1, while lz4hc is configured as a compression algorithm with priority 2.
461 Alternative compression algorithm's priority i 436 Alternative compression algorithm's priority is provided during algorithms
462 configuration::: 437 configuration:::
463 438
464 #select zstd recompression algorithm, 439 #select zstd recompression algorithm, priority 1
465 echo "algo=zstd priority=1" > /sys/blo 440 echo "algo=zstd priority=1" > /sys/block/zramX/recomp_algorithm
466 441
467 #select deflate recompression algorith 442 #select deflate recompression algorithm, priority 2
468 echo "algo=deflate priority=2" > /sys/ 443 echo "algo=deflate priority=2" > /sys/block/zramX/recomp_algorithm
469 444
470 Another device attribute that CONFIG_ZRAM_MULT 445 Another device attribute that CONFIG_ZRAM_MULTI_COMP enables is recompress,
471 which controls recompression. 446 which controls recompression.
472 447
473 Examples::: 448 Examples:::
474 449
475 #IDLE pages recompression is activated 450 #IDLE pages recompression is activated by `idle` mode
476 echo "type=idle" > /sys/block/zramX/re 451 echo "type=idle" > /sys/block/zramX/recompress
477 452
478 #HUGE pages recompression is activated 453 #HUGE pages recompression is activated by `huge` mode
479 echo "type=huge" > /sys/block/zram0/re 454 echo "type=huge" > /sys/block/zram0/recompress
480 455
481 #HUGE_IDLE pages recompression is acti 456 #HUGE_IDLE pages recompression is activated by `huge_idle` mode
482 echo "type=huge_idle" > /sys/block/zra 457 echo "type=huge_idle" > /sys/block/zramX/recompress
483 458
484 The number of idle pages can be significant, s 459 The number of idle pages can be significant, so user-space can pass a size
485 threshold (in bytes) to the recompress knob: z 460 threshold (in bytes) to the recompress knob: zram will recompress only pages
486 of equal or greater size::: 461 of equal or greater size:::
487 462
488 #recompress all pages larger than 3000 463 #recompress all pages larger than 3000 bytes
489 echo "threshold=3000" > /sys/block/zra 464 echo "threshold=3000" > /sys/block/zramX/recompress
490 465
491 #recompress idle pages larger than 200 466 #recompress idle pages larger than 2000 bytes
492 echo "type=idle threshold=2000" > /sys 467 echo "type=idle threshold=2000" > /sys/block/zramX/recompress
493 468
494 It is also possible to limit the number of pag <<
495 attempt to recompress::: <<
496 <<
497 echo "type=huge_idle max_pages=42" > / <<
498 <<
499 Recompression of idle pages requires memory tr 469 Recompression of idle pages requires memory tracking.
500 470
501 During re-compression for every page, that mat 471 During re-compression for every page, that matches re-compression criteria,
502 ZRAM iterates the list of registered alternati 472 ZRAM iterates the list of registered alternative compression algorithms in
503 order of their priorities. ZRAM stops either w 473 order of their priorities. ZRAM stops either when re-compression was
504 successful (re-compressed object is smaller in 474 successful (re-compressed object is smaller in size than the original one)
505 and matches re-compression criteria (e.g. size 475 and matches re-compression criteria (e.g. size threshold) or when there are
506 no secondary algorithms left to try. If none o 476 no secondary algorithms left to try. If none of the secondary algorithms can
507 successfully re-compressed the page such a pag 477 successfully re-compressed the page such a page is marked as incompressible,
508 so ZRAM will not attempt to re-compress it in 478 so ZRAM will not attempt to re-compress it in the future.
509 479
510 This re-compression behaviour, when it iterate 480 This re-compression behaviour, when it iterates through the list of
511 registered compression algorithms, increases o 481 registered compression algorithms, increases our chances of finding the
512 algorithm that successfully compresses a parti 482 algorithm that successfully compresses a particular page. Sometimes, however,
513 it is convenient (and sometimes even necessary 483 it is convenient (and sometimes even necessary) to limit recompression to
514 only one particular algorithm so that it will 484 only one particular algorithm so that it will not try any other algorithms.
515 This can be achieved by providing a `algo` or !! 485 This can be achieved by providing a algo=NAME parameter:::
516 486
517 #use zstd algorithm only (if registere 487 #use zstd algorithm only (if registered)
518 echo "type=huge algo=zstd" > /sys/bloc 488 echo "type=huge algo=zstd" > /sys/block/zramX/recompress
519 <<
520 #use zstd algorithm only (if zstd was <<
521 echo "type=huge priority=1" > /sys/blo <<
522 489
523 memory tracking 490 memory tracking
524 =============== 491 ===============
525 492
526 With CONFIG_ZRAM_MEMORY_TRACKING, user can kno 493 With CONFIG_ZRAM_MEMORY_TRACKING, user can know information of the
527 zram block. It could be useful to catch cold o 494 zram block. It could be useful to catch cold or incompressible
528 pages of the process with*pagemap. 495 pages of the process with*pagemap.
529 496
530 If you enable the feature, you could see block 497 If you enable the feature, you could see block state via
531 /sys/kernel/debug/zram/zram0/block_state". The 498 /sys/kernel/debug/zram/zram0/block_state". The output is as follows::
532 499
533 300 75.033841 .wh... 500 300 75.033841 .wh...
534 301 63.806904 s..... 501 301 63.806904 s.....
535 302 63.806919 ..hi.. 502 302 63.806919 ..hi..
536 303 62.801919 ....r. 503 303 62.801919 ....r.
537 304 146.781902 ..hi.n 504 304 146.781902 ..hi.n
538 505
539 First column 506 First column
540 zram's block index. 507 zram's block index.
541 Second column 508 Second column
542 access time since the system was boote 509 access time since the system was booted
543 Third column 510 Third column
544 state of the block: 511 state of the block:
545 512
546 s: 513 s:
547 same page 514 same page
548 w: 515 w:
549 written page to backing store 516 written page to backing store
550 h: 517 h:
551 huge page 518 huge page
552 i: 519 i:
553 idle page 520 idle page
554 r: 521 r:
555 recompressed page (secondary c 522 recompressed page (secondary compression algorithm)
556 n: 523 n:
557 none (including secondary) of 524 none (including secondary) of algorithms could compress it
558 525
559 First line of above example says 300th block i 526 First line of above example says 300th block is accessed at 75.033841sec
560 and the block's state is huge so it is written 527 and the block's state is huge so it is written back to the backing
561 storage. It's a debugging feature so anyone sh 528 storage. It's a debugging feature so anyone shouldn't rely on it to work
562 properly. 529 properly.
563 530
564 Nitin Gupta 531 Nitin Gupta
565 ngupta@vflare.org 532 ngupta@vflare.org
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.