1 zstd(1) -- zstd, zstdmt, unzstd, zstdcat - Compress or decompress .zst files
2 ============================================================================
7 `zstd` [*OPTIONS*] [-|_INPUT-FILE_] [-o _OUTPUT-FILE_]
9 `zstdmt` is equivalent to `zstd -T0`
11 `unzstd` is equivalent to `zstd -d`
13 `zstdcat` is equivalent to `zstd -dcf`
18 `zstd` is a fast lossless compression algorithm and data compression tool,
19 with command line syntax similar to `gzip (1)` and `xz (1)`.
20 It is based on the **LZ77** family, with further FSE & huff0 entropy stages.
21 `zstd` offers highly configurable compression speed,
22 with fast modes at > 200 MB/s per code,
23 and strong modes nearing lzma compression ratios.
24 It also features a very fast decoder, with speeds > 500 MB/s per core.
26 `zstd` command line syntax is generally similar to gzip,
27 but features the following differences :
29 - Source files are preserved by default.
30 It's possible to remove them automatically by using the `--rm` command.
31 - When compressing a single file, `zstd` displays progress notifications
32 and result summary by default.
33 Use `-q` to turn them off.
34 - `zstd` does not accept input from console,
35 but it properly accepts `stdin` when it's not the console.
36 - `zstd` displays a short help page when command line is an error.
37 Use `-q` to turn it off.
39 `zstd` compresses or decompresses each _file_ according to the selected
41 If no _files_ are given or _file_ is `-`, `zstd` reads from standard input
42 and writes the processed data to standard output.
43 `zstd` will refuse to write compressed data to standard output
44 if it is a terminal : it will display an error message and skip the _file_.
45 Similarly, `zstd` will refuse to read compressed data from standard input
48 Unless `--stdout` or `-o` is specified, _files_ are written to a new file
49 whose name is derived from the source _file_ name:
51 * When compressing, the suffix `.zst` is appended to the source filename to
52 get the target filename.
53 * When decompressing, the `.zst` suffix is removed from the source filename to
54 get the target filename
56 ### Concatenation with .zst files
57 It is possible to concatenate `.zst` files as is.
58 `zstd` will decompress such files as if they were a single `.zst` file.
63 ### Integer suffixes and special values
64 In most places where an integer argument is expected,
65 an optional suffix is supported to easily indicate large integers.
66 There must be no space between the integer and the suffix.
69 Multiply the integer by 1,024 (2\^10).
70 `Ki`, `K`, and `KB` are accepted as synonyms for `KiB`.
72 Multiply the integer by 1,048,576 (2\^20).
73 `Mi`, `M`, and `MB` are accepted as synonyms for `MiB`.
76 If multiple operation mode options are given,
77 the last one takes effect.
81 This is the default operation mode when no operation mode option is specified
82 and no other operation mode is implied from the command name
83 (for example, `unzstd` implies `--decompress`).
84 * `-d`, `--decompress`, `--uncompress`:
87 Test the integrity of compressed _files_.
88 This option is equivalent to `--decompress --stdout` except that the
89 decompressed data is discarded instead of being written to standard output.
90 No files are created or removed.
92 Benchmark file(s) using compression level #
94 Use FILEs as a training set to create a dictionary.
95 The training set should contain a lot of small files (> 100).
97 Display information related to a zstd compressed file, such as size, ratio, and checksum.
98 Some of these fields may not be available.
99 This command can be augmented with the `-v` modifier.
101 ### Operation modifiers
104 `#` compression level \[1-19] (default: 3)
106 unlocks high compression levels 20+ (maximum 22), using a lot more memory.
107 Note that decompression will also require more memory when using these levels.
109 enables long distance matching with `#` `windowLog`, if not `#` is not
110 present it defaults to `27`.
111 This increases the window size (`windowLog`) and memory usage for both the
112 compressor and decompressor.
113 This setting is designed to improve the compression ratio for files with
114 long matches at a large distance.
116 Note: If `windowLog` is set to larger than 27, `--long=windowLog` or
117 `--memory=windowSize` needs to be passed to the decompressor.
118 * `-T#`, `--threads=#`:
119 Compress using `#` threads (default: 1).
120 If `#` is 0, attempt to detect and use the number of physical CPU cores.
121 In all cases, the nb of threads is capped to ZSTDMT_NBTHREADS_MAX==256.
122 This modifier does nothing if `zstd` is compiled without multithread support.
124 use `file` as Dictionary to compress or decompress FILE(s)
126 do not store dictionary ID within frame header (dictionary compression).
127 The decoder will have to rely on implicit knowledge about which dictionary to use,
128 it won't be able to check if it's correct.
130 save result into `file` (only possible with a single _INPUT-FILE_)
132 overwrite output without prompting, and (de)compress symbolic links
134 force write to standard output, even if it is the console
136 enable / disable sparse FS support,
137 to make files with many zeroes smaller on disk.
138 Creating sparse files may save disk space and speed up decompression by
139 reducing the amount of disk I/O.
140 default : enabled when output is into a file,
141 and disabled when output is stdout.
142 This setting overrides default and can force sparse mode over stdout.
144 remove source file(s) after successful compression or decompression
146 keep source file(s) after successful compression or decompression.
147 This is the default behavior.
149 operate recursively on dictionaries
151 compress and decompress in other formats. If compiled with
152 support, zstd can compress to or decompress from other compression algorithm
153 formats. Possibly available options are `gzip`, `xz`, `lzma`, and `lz4`.
154 * `-h`/`-H`, `--help`:
155 display help/long help and exit
157 display version number and exit.
158 Advanced : `-vV` also displays supported formats.
159 `-vvV` also displays POSIX support.
163 suppress warnings, interactivity, and notifications.
164 specify twice to suppress errors too.
165 * `-C`, `--[no-]check`:
166 add integrity check computed from uncompressed data (default : enabled)
168 All arguments after `--` are treated as files
173 `zstd` offers _dictionary_ compression,
174 useful for very small files and messages.
175 It's possible to train `zstd` with some samples,
176 the result of which is saved into a file called a `dictionary`.
177 Then during compression and decompression, reference the same dictionary.
178 It will improve compression ratio of small files.
179 Typical gains range from 10% (at 64KB) to x5 better (at <1KB).
182 Use FILEs as training set to create a dictionary.
183 The training set should contain a lot of small files (> 100),
184 and weight typically 100x the target dictionary size
185 (for example, 10 MB for a 100 KB dictionary).
187 Supports multithreading if `zstd` is compiled with threading support.
188 Additional parameters can be specified with `--train-cover`.
189 The legacy dictionary builder can be accessed with `--train-legacy`.
190 Equivalent to `--train-cover=d=8,steps=4`.
192 Dictionary saved into `file` (default name: dictionary).
194 Limit dictionary to specified size (default: 112640).
196 Split input files in blocks of size # (default: no split)
198 A dictionary ID is a locally unique ID that a decoder can use to verify it is
199 using the right dictionary.
200 By default, zstd will create a 4-bytes random number ID.
201 It's possible to give a precise number instead.
202 Short numbers have an advantage : an ID < 256 will only need 1 byte in the
203 compressed frame header, and an ID < 65536 will only need 2 bytes.
204 This compares favorably to 4 bytes default.
205 However, it's up to the dictionary manager to not assign twice the same ID to
206 2 different dictionaries.
207 * `--train-cover[=k#,d=#,steps=#]`:
208 Select parameters for the default dictionary builder algorithm named cover.
209 If _d_ is not specified, then it tries _d_ = 6 and _d_ = 8.
210 If _k_ is not specified, then it tries _steps_ values in the range [50, 2000].
211 If _steps_ is not specified, then the default value of 40 is used.
212 Requires that _d_ <= _k_.
214 Selects segments of size _k_ with highest score to put in the dictionary.
215 The score of a segment is computed by the sum of the frequencies of all the
216 subsegments of size _d_.
217 Generally _d_ should be in the range [6, 8], occasionally up to 16, but the
218 algorithm will run faster with d <= _8_.
219 Good values for _k_ vary widely based on the input data, but a safe range is
221 Supports multithreading if `zstd` is compiled with threading support.
225 `zstd --train-cover FILEs`
227 `zstd --train-cover=k=50,d=8 FILEs`
229 `zstd --train-cover=d=8,steps=500 FILEs`
231 `zstd --train-cover=k=50 FILEs`
233 * `--train-legacy[=selectivity=#]`:
234 Use legacy dictionary builder algorithm with the given dictionary
235 _selectivity_ (default: 9).
236 The smaller the _selectivity_ value, the denser the dictionary,
237 improving its efficiency but reducing its possible maximum size.
238 `--train-legacy=s=#` is also accepted.
242 `zstd --train-legacy FILEs`
244 `zstd --train-legacy=selectivity=8 FILEs`
251 benchmark file(s) using compression level #
253 benchmark file(s) using multiple compression levels, from `-b#` to `-e#` (inclusive)
255 minimum evaluation time, in seconds (default : 3s), benchmark mode only
256 * `-B#`, `--block-size=#`:
257 cut file(s) into independent blocks of size # (default: no block)
259 set process priority to real-time
262 ADVANCED COMPRESSION OPTIONS
263 ----------------------------
264 ### --zstd[=options]:
265 `zstd` provides 22 predefined compression levels.
266 The selected or default predefined compression level can be changed with
267 advanced compression options.
268 The _options_ are provided as a comma-separated list.
269 You may specify only the options you want to change and the rest will be
270 taken from the selected or default compression level.
271 The list of available _options_:
273 - `strategy`=_strat_, `strat`=_strat_:
274 Specify a strategy used by a match finder.
276 There are 8 strategies numbered from 1 to 8, from faster to stronger:
277 1=ZSTD\_fast, 2=ZSTD\_dfast, 3=ZSTD\_greedy, 4=ZSTD\_lazy,
278 5=ZSTD\_lazy2, 6=ZSTD\_btlazy2, 7=ZSTD\_btopt, 8=ZSTD\_btultra.
280 - `windowLog`=_wlog_, `wlog`=_wlog_:
281 Specify the maximum number of bits for a match distance.
283 The higher number of increases the chance to find a match which usually
284 improves compression ratio.
285 It also increases memory requirements for the compressor and decompressor.
286 The minimum _wlog_ is 10 (1 KiB) and the maximum is 30 (1 GiB) on 32-bit
287 platforms and 31 (2 GiB) on 64-bit platforms.
289 Note: If `windowLog` is set to larger than 27, `--long=windowLog` or
290 `--memory=windowSize` needs to be passed to the decompressor.
292 - `hashLog`=_hlog_, `hlog`=_hlog_:
293 Specify the maximum number of bits for a hash table.
295 Bigger hash tables cause less collisions which usually makes compression
296 faster, but requires more memory during compression.
298 The minimum _hlog_ is 6 (64 B) and the maximum is 26 (128 MiB).
300 - `chainLog`=_clog_, `clog`=_clog_:
301 Specify the maximum number of bits for a hash chain or a binary tree.
303 Higher numbers of bits increases the chance to find a match which usually
304 improves compression ratio.
305 It also slows down compression speed and increases memory requirements for
307 This option is ignored for the ZSTD_fast strategy.
309 The minimum _clog_ is 6 (64 B) and the maximum is 28 (256 MiB).
311 - `searchLog`=_slog_, `slog`=_slog_:
312 Specify the maximum number of searches in a hash chain or a binary tree
313 using logarithmic scale.
315 More searches increases the chance to find a match which usually increases
316 compression ratio but decreases compression speed.
318 The minimum _slog_ is 1 and the maximum is 26.
320 - `searchLength`=_slen_, `slen`=_slen_:
321 Specify the minimum searched length of a match in a hash table.
323 Larger search lengths usually decrease compression ratio but improve
326 The minimum _slen_ is 3 and the maximum is 7.
328 - `targetLen`=_tlen_, `tlen`=_tlen_:
329 Specify the minimum match length that causes a match finder to stop
330 searching for better matches.
332 A larger minimum match length usually improves compression ratio but
333 decreases compression speed.
334 This option is only used with strategies ZSTD_btopt and ZSTD_btultra.
336 The minimum _tlen_ is 4 and the maximum is 999.
338 - `overlapLog`=_ovlog_, `ovlog`=_ovlog_:
339 Determine `overlapSize`, amount of data reloaded from previous job.
340 This parameter is only available when multithreading is enabled.
341 Reloading more data improves compression ratio, but decreases speed.
343 The minimum _ovlog_ is 0, and the maximum is 9.
344 0 means "no overlap", hence completely independent jobs.
345 9 means "full overlap", meaning up to `windowSize` is reloaded from previous job.
346 Reducing _ovlog_ by 1 reduces the amount of reload by a factor 2.
347 Default _ovlog_ is 6, which means "reload `windowSize / 8`".
348 Exception : the maximum compression level (22) has a default _ovlog_ of 9.
350 - `ldmHashLog`=_ldmhlog_, `ldmhlog`=_ldmhlog_:
351 Specify the maximum size for a hash table used for long distance matching.
353 This option is ignored unless long distance matching is enabled.
355 Bigger hash tables usually improve compression ratio at the expense of more
356 memory during compression and a decrease in compression speed.
358 The minimum _ldmhlog_ is 6 and the maximum is 26 (default: 20).
360 - `ldmSearchLength`=_ldmslen_, `ldmslen`=_ldmslen_:
361 Specify the minimum searched length of a match for long distance matching.
363 This option is ignored unless long distance matching is enabled.
365 Larger/very small values usually decrease compression ratio.
367 The minumum _ldmslen_ is 4 and the maximum is 4096 (default: 64).
369 - `ldmBucketSizeLog`=_ldmblog_, `ldmblog`=_ldmblog_:
370 Specify the size of each bucket for the hash table used for long distance
373 This option is ignored unless long distance matching is enabled.
375 Larger bucket sizes improve collision resolution but decrease compression
378 The minimum _ldmblog_ is 0 and the maximum is 8 (default: 3).
380 - `ldmHashEveryLog`=_ldmhevery_, `ldmhevery`=_ldmhevery_:
381 Specify the frequency of inserting entries into the long distance matching
384 This option is ignored unless long distance matching is enabled.
386 Larger values will improve compression speed. Deviating far from the
387 default value will likely result in a decrease in compression ratio.
389 The default value is `wlog - ldmhlog`.
392 Select the size of each compression job.
393 This parameter is available only when multi-threading is enabled.
394 Default value is `4 * windowSize`, which means it varies depending on compression level.
395 `-B#` makes it possible to select a custom value.
396 Note that job size must respect a minimum value which is enforced transparently.
397 This minimum is either 1 MB, or `overlapSize`, whichever is largest.
400 The following parameters sets advanced compression options to those of
401 predefined level 19 for files bigger than 256 KB:
403 `--zstd`=windowLog=23,chainLog=23,hashLog=22,searchLog=6,searchLength=3,targetLength=48,strategy=6
407 Report bugs at: https://github.com/facebook/zstd/issues