3 .\" Author: Lasse Collin
5 .\" This file has been put into the public domain.
6 .\" You can do whatever you want with this file.
8 .TH XZ 1 "2010-03-07" "Tukaani" "XZ Utils"
10 xz, unxz, xzcat, lzma, unlzma, lzcat \- Compress or decompress .xz and .lzma files
18 .BR "xz \-\-decompress" .
22 .BR "xz \-\-decompress \-\-stdout" .
26 .BR "xz \-\-format=lzma" .
30 .BR "xz \-\-format=lzma \-\-decompress" .
34 .BR "xz \-\-format=lzma \-\-decompress \-\-stdout" .
36 When writing scripts that need to decompress files, it is recommended to
39 with appropriate arguments
49 is a general-purpose data compression tool with command line syntax similar to
53 The native file format is the
55 format, but also the legacy
57 format and raw compressed streams with no container format headers
61 compresses or decompresses each
63 according to the selected operation mode.
71 reads from standard input and writes the processed data to standard output.
73 will refuse (display an error and skip the
75 to write compressed data to standard output if it is a terminal. Similarly,
77 will refuse to read compressed data from standard input if it is a terminal.
85 are written to a new file whose name is derived from the source
89 When compressing, the suffix of the target file format
93 is appended to the source filename to get the target filename.
95 When decompressing, the
99 suffix is removed from the filename to get the target filename.
101 also recognizes the suffixes
105 and replaces them with the
109 If the target file already exists, an error is displayed and the
113 Unless writing to standard output,
115 will display a warning and skip the
117 if any of the following applies:
120 is not a regular file. Symbolic links are not followed, thus they
121 are not considered to be regular files.
124 has more than one hard link.
127 has setuid, setgid, or sticky bit set.
129 The operation mode is set to compress, and the
131 already has a suffix of the target file format
135 when compressing to the
141 when compressing to the
145 The operation mode is set to decompress, and the
147 doesn't have a suffix of any of the supported file formats
154 After successfully compressing or decompressing the
157 copies the owner, group, permissions, access time, and modification time
160 to the target file. If copying the group fails, the permissions are modified
161 so that the target file doesn't become accessible to users who didn't have
162 permission to access the source
165 doesn't support copying other metadata like access control lists
166 or extended attributes yet.
168 Once the target file has been successfully closed, the source
172 was specified. The source
174 is never removed if the output is written to standard output.
182 process makes it print progress information to standard error.
183 This has only limited use since when standard error is a terminal, using
185 will display an automatically updating progress indicator.
189 varies from a few hundred kilobytes to several gigabytes depending on
190 the compression settings. The settings used when compressing a file
191 affect also the memory usage of the decompressor. Typically the decompressor
192 needs only 5\ % to 20\ % of the amount of RAM that the compressor needed when
193 creating the file. Still, the worst-case memory usage of the decompressor
194 is several gigabytes.
196 To prevent uncomfortable surprises caused by huge memory usage,
198 has a built-in memory usage limiter. While some operating systems provide
199 ways to limit the memory usage of processes, relying on it wasn't deemed
200 to be flexible enough. The default limit depends on the total amount of
203 If 40\ % of RAM is at least 80 MiB, 40\ % of RAM is used as the limit.
205 If 80\ % of RAM is over 80 MiB, 80 MiB is used as the limit.
207 Otherwise 80\ % of RAM is used as the limit.
209 When compressing, if the selected compression settings exceed the memory
210 usage limit, the settings are automatically adjusted downwards and a notice
211 about this is displayed. As an exception, if the memory usage limit is
212 exceeded when compressing with
214 an error is displayed and
216 will exit with exit status
221 cannot be decompressed without exceeding the memory usage limit, an error
222 message is displayed and the file is skipped. Note that compressed files
223 may contain many blocks, which may have been compressed with different
224 settings. Typically all blocks will have roughly the same memory requirements,
225 but it is possible that a block later in the file will exceed the memory usage
226 limit, and an error about too low memory usage limit gets displayed after some
227 data has already been decompressed.
229 The absolute value of the active memory usage limit can be seen with
231 or near the bottom of the output of
233 The default limit can be overridden with
234 \fB\-\-memory=\fIlimit\fR.
236 .SS "Integer suffixes and special values"
237 In most places where an integer argument is expected, an optional suffix
238 is supported to easily indicate large integers. There must be no space
239 between the integer and the suffix.
242 The integer is multiplied by 1,024 (2^10). Also
249 are accepted as synonyms for
253 The integer is multiplied by 1,048,576 (2^20). Also
259 are accepted as synonyms for
263 The integer is multiplied by 1,073,741,824 (2^30). Also
269 are accepted as synonyms for
274 can be used to indicate the maximum integer value supported by the option.
276 If multiple operation mode options are given, the last one takes effect.
278 .BR \-z ", " \-\-compress
279 Compress. This is the default operation mode when no operation mode option
280 is specified, and no other operation mode is implied from the command name
284 .BR \-\-decompress ).
286 .BR \-d ", " \-\-decompress ", " \-\-uncompress
289 .BR \-t ", " \-\-test
290 Test the integrity of compressed
292 No files are created or removed. This option is equivalent to
293 .B "\-\-decompress \-\-stdout"
294 except that the decompressed data is discarded instead of being
295 written to standard output.
297 .BR \-l ", " \-\-list
298 View information about the compressed files. No uncompressed output is
299 produced, and no files are created or removed. In list mode, the program
300 cannot read the compressed data from standard input or from other
303 .B "This feature has not been implemented yet."
304 .SS "Operation modifiers"
306 .BR \-k ", " \-\-keep
307 Keep (don't delete) the input files.
309 .BR \-f ", " \-\-force
310 This option has several effects:
313 If the target file already exists, delete it before compressing or
316 Compress or decompress even if the input is a symbolic link to a regular file,
317 has more than one hard link, or has setuid, setgid, or sticky bit set.
318 The setuid, setgid, and sticky bits are not copied to the target file.
325 doesn't recognize the type of the source file,
327 will copy the source file as is to standard output. This allows using
332 for files that have not been compressed with
336 might support new compressed file formats, which may make
338 decompress more types of files instead of copying them as is to
340 .BI \-\-format= format
341 can be used to restrict
343 to decompress only a single file format.
346 .BR \-c ", " \-\-stdout ", " \-\-to-stdout
347 Write the compressed or decompressed data to standard output instead of
352 Disable creation of sparse files. By default, if decompressing into
355 tries to make the file sparse if the decompressed data contains long
356 sequences of binary zeros. It works also when writing to standard output
357 as long as standard output is connected to a regular file, and certain
358 additional conditions are met to make it safe. Creating sparse files may
359 save disk space and speed up the decompression by reducing the amount of
362 \fB\-S\fR \fI.suf\fR, \fB\-\-suffix=\fI.suf
363 When compressing, use
365 as the suffix for the target file instead of
369 If not writing to standard output and the source file already has the suffix
371 a warning is displayed and the file is skipped.
373 When decompressing, recognize also files with the suffix
375 in addition to files with the
381 suffix. If the source file has the suffix
383 the suffix is removed to get the target filename.
385 When compressing or decompressing raw streams
386 .RB ( \-\-format=raw ),
387 the suffix must always be specified unless writing to standard output,
388 because there is no default suffix for raw streams.
390 \fB\-\-files\fR[\fB=\fIfile\fR]
391 Read the filenames to process from
395 is omitted, filenames are read from standard input. Filenames must be
396 terminated with the newline character. A dash
398 is taken as a regular filename; it doesn't mean standard input.
399 If filenames are given also as command line arguments, they are
400 processed before the filenames read from
403 \fB\-\-files0\fR[\fB=\fIfile\fR]
404 This is identical to \fB\-\-files\fR[\fB=\fIfile\fR] except that the
405 filenames must be terminated with the null character.
406 .SS "Basic file format and compression options"
408 \fB\-F\fR \fIformat\fR, \fB\-\-format=\fIformat
409 Specify the file format to compress or decompress:
413 This is the default. When compressing,
417 When decompressing, the format of the input file is automatically detected.
418 Note that raw streams (created with
420 cannot be auto-detected.
425 file format, or accept only
427 files when decompressing.
432 Compress to the legacy
434 file format, or accept only
436 files when decompressing. The alternative name
438 is provided for backwards compatibility with LZMA Utils.
441 Compress or uncompress a raw stream (no headers). This is meant for advanced
442 users only. To decode raw streams, you need to set not only
444 but also specify the filter chain, which would normally be stored in the
445 container format headers.
448 \fB\-C\fR \fIcheck\fR, \fB\-\-check=\fIcheck
449 Specify the type of the integrity check, which is calculated from the
450 uncompressed data. This option has an effect only when compressing into the
454 format doesn't support integrity checks.
455 The integrity check (if any) is verified when the
457 file is decompressed.
465 Don't calculate an integrity check at all. This is usually a bad idea. This
466 can be useful when integrity of the data is verified by other means anyway.
469 Calculate CRC32 using the polynomial from IEEE-802.3 (Ethernet).
472 Calculate CRC64 using the polynomial from ECMA-182. This is the default, since
473 it is slightly better than CRC32 at detecting damaged files and the speed
474 difference is negligible.
477 Calculate SHA-256. This is somewhat slower than CRC32 and CRC64.
482 headers is always verified with CRC32. It is not possible to change or
486 Select compression preset. If a preset level is specified multiple times,
487 the last one takes effect.
489 The compression preset levels can be categorised roughly into three
492 .IP "\fB\-0\fR ... \fB\-2"
493 Fast presets with relatively low memory usage.
497 should give compression speed and ratios comparable to
504 is not very good (not much faster than
506 but much worse compression). In future,
508 may be indicate some fast algorithm instead of LZMA2.
509 .IP "\fB\-3\fR ... \fB\-5"
510 Good compression ratio with low to medium memory usage.
511 These are significantly slower than levels 0\-2.
512 .IP "\fB\-6\fR ... \fB\-9"
513 Excellent compression with medium to high memory usage. These are also
514 slower than the lower preset levels. The default is
516 Unless you want to maximize the compression ratio, you probably don't want
517 a higher preset level than
519 due to speed and memory usage.
522 The exact compression settings (filter chain) used by each preset may
525 versions. The settings may also vary between files being compressed, if
527 determines that modified settings will probably give better compression
528 ratio without significantly affecting compression time or memory usage.
530 Because the settings may vary, the memory usage may vary too. The following
531 table lists the maximum memory usage of each preset level, which won't be
532 exceeded even in future versions of
535 .B "FIXME: The table below is just a rough idea."
542 Preset;Compression;Decompression
559 automatically adjusts the compression settings downwards if
560 the memory usage limit would be exceeded, so it is safe to specify
561 a high preset level even on systems that don't have lots of RAM.
563 .BR \-\-fast " and " \-\-best
564 These are somewhat misleading aliases for
569 These are provided only for backwards compatibility with LZMA Utils.
570 Avoid using these options.
572 Especially the name of
574 is misleading, because the definition of best depends on the input data,
575 and that usually people don't want the very best compression ratio anyway,
576 because it would be very slow.
578 .BR \-e ", " \-\-extreme
579 Modify the compression preset (\fB\-0\fR ... \fB\-9\fR) so that a little bit
580 better compression ratio can be achieved without increasing memory usage
581 of the compressor or decompressor (exception: compressor memory usage may
582 increase a little with presets \fB\-0\fR ... \fB\-2\fR). The downside is that
583 the compression time will increase dramatically (it can easily double).
585 \fB\-M\fR \fIlimit\fR, \fB\-\-memory=\fIlimit
586 Set the memory usage limit. If this option is specified multiple times,
587 the last one takes effect. The
589 can be specified in multiple ways:
594 can be an absolute value in bytes. Using an integer suffix like
596 can be useful. Example:
597 .B "\-\-memory=80MiB"
601 can be specified as a percentage of physical RAM. Example:
606 can be reset back to its default value by setting it to
610 for how the default limit is defined.
612 The memory usage limiting can be effectively disabled by setting
616 This isn't recommended. It's usually better to use, for example,
622 can be seen near the bottom of the output of the
626 \fB\-T\fR \fIthreads\fR, \fB\-\-threads=\fIthreads
627 Specify the maximum number of worker threads to use. The default is
628 the number of available CPU cores. You can see the current value of
630 near the end of the output of the
634 The actual number of worker threads can be less than
636 if using more threads would exceed the memory usage limit.
637 In addition to CPU-intensive worker threads,
639 may use a few auxiliary threads, which don't use a lot of CPU time.
641 .B "Multithreaded compression and decompression are not implemented yet,"
642 .B "so this option has no effect for now."
643 .SS Custom compressor filter chains
644 A custom filter chain allows specifying the compression settings in detail
645 instead of relying on the settings associated to the preset levels.
646 When a custom filter chain is specified, the compression preset level options
647 (\fB\-0\fR ... \fB\-9\fR and \fB\-\-extreme\fR) are silently ignored.
649 A filter chain is comparable to piping on the UN*X command line.
650 When compressing, the uncompressed input goes to the first filter, whose
651 output goes to the next filter (if any). The output of the last filter
652 gets written to the compressed file. The maximum number of filters in
653 the chain is four, but typically a filter chain has only one or two filters.
655 Many filters have limitations where they can be in the filter chain:
656 some filters can work only as the last filter in the chain, some only
657 as a non-last filter, and some work in any position in the chain. Depending
658 on the filter, this limitation is either inherent to the filter design or
659 exists to prevent security issues.
661 A custom filter chain is specified by using one or more filter options in
662 the order they are wanted in the filter chain. That is, the order of filter
663 options is significant! When decoding raw streams
664 .RB ( \-\-format=raw ),
665 the filter chain is specified in the same order as it was specified when
668 Filters take filter-specific
670 as a comma-separated list. Extra commas in
672 are ignored. Every option has a default value, so you need to
673 specify only those you want to change.
675 \fB\-\-lzma1\fR[\fB=\fIoptions\fR], \fB\-\-lzma2\fR[\fB=\fIoptions\fR]
676 Add LZMA1 or LZMA2 filter to the filter chain. These filter can be used
677 only as the last filter in the chain.
679 LZMA1 is a legacy filter, which is supported almost solely due to the legacy
681 file format, which supports only LZMA1. LZMA2 is an updated
682 version of LZMA1 to fix some practical issues of LZMA1. The
684 format uses LZMA2, and doesn't support LZMA1 at all. Compression speed and
685 ratios of LZMA1 and LZMA2 are practically the same.
687 LZMA1 and LZMA2 share the same set of
692 Reset all LZMA1 or LZMA2
697 consist of an integer, which may be followed by single-letter preset
698 modifiers. The integer can be from
702 matching the command line options \fB\-0\fR ... \fB\-9\fR.
703 The only supported modifier is currently
712 from which the default values for the rest of the LZMA1 or LZMA2
717 Dictionary (history buffer) size indicates how many bytes of the recently
718 processed uncompressed data is kept in memory. One method to reduce size of
719 the uncompressed data is to store distance-length pairs, which
720 indicate what data to repeat from the dictionary buffer. The bigger
721 the dictionary, the better the compression ratio usually is,
722 but dictionaries bigger than the uncompressed data are waste of RAM.
724 Typical dictionary size is from 64 KiB to 64 MiB. The minimum is 4 KiB.
725 The maximum for compression is currently 1.5 GiB. The decompressor already
726 supports dictionaries up to one byte less than 4 GiB, which is the
727 maximum for LZMA1 and LZMA2 stream formats.
729 Dictionary size has the biggest effect on compression ratio.
730 Dictionary size and match finder together determine the memory usage of
731 the LZMA1 or LZMA2 encoder. The same dictionary size is required
732 for decompressing that was used when compressing, thus the memory usage of
733 the decoder is determined by the dictionary size used when compressing.
736 Specify the number of literal context bits. The minimum is
742 In addition, the sum of
750 Specify the number of literal position bits. The minimum is
758 Specify the number of position bits. The minimum is
768 specifies the function used to analyze the data produced by the match finder.
787 Match finder has a major effect on encoder speed, memory usage, and
788 compression ratio. Usually Hash Chain match finders are faster than
789 Binary Tree match finders. Hash Chains are usually used together with
791 and Binary Trees with
793 The memory usage formulas are only rough estimates,
794 which are closest to reality when
800 Hash Chain with 2- and 3-byte hashing
818 Hash Chain with 2-, 3-, and 4-byte hashing
829 Binary Tree with 2-byte hashing
840 Binary Tree with 2- and 3-byte hashing
858 Binary Tree with 2-, 3-, and 4-byte hashing
870 Specify what is considered to be a nice length for a match. Once a match
873 bytes is found, the algorithm stops looking for possibly better matches.
876 can be 2\-273 bytes. Higher values tend to give better compression ratio
877 at expense of speed. The default depends on the
882 Specify the maximum search depth in the match finder. The default is the
885 which makes the compressor determine a reasonable
892 Using very high values for
894 can make the encoder extremely slow with carefully crafted files.
897 over 1000 unless you are prepared to interrupt the compression in case it
901 When decoding raw streams
902 .RB ( \-\-format=raw ),
903 LZMA2 needs only the value of
911 \fB\-\-x86\fR[\fB=\fIoptions\fR]
913 \fB\-\-powerpc\fR[\fB=\fIoptions\fR]
915 \fB\-\-ia64\fR[\fB=\fIoptions\fR]
917 \fB\-\-arm\fR[\fB=\fIoptions\fR]
919 \fB\-\-armthumb\fR[\fB=\fIoptions\fR]
921 \fB\-\-sparc\fR[\fB=\fIoptions\fR]
922 Add a branch/call/jump (BCJ) filter to the filter chain. These filters
923 can be used only as non-last filter in the filter chain.
925 A BCJ filter converts relative addresses in the machine code to their
926 absolute counterparts. This doesn't change the size of the data, but
927 it increases redundancy, which allows e.g. LZMA2 to get better
930 The BCJ filters are always reversible, so using a BCJ filter for wrong
931 type of data doesn't cause any data loss. However, applying a BCJ filter
932 for wrong type of data is a bad idea, because it tends to make the
933 compression ratio worse.
935 Different instruction sets have have different alignment:
942 Filter;Alignment;Notes
943 x86;1;32-bit and 64-bit x86
944 PowerPC;4;Big endian only
945 ARM;4;Little endian only
946 ARM-Thumb;2;Little endian only
947 IA-64;16;Big or little endian
948 SPARC;4;Big or little endian
953 Since the BCJ-filtered data is usually compressed with LZMA2, the compression
954 ratio may be improved slightly if the LZMA2 options are set to match the
955 alignment of the selected BCJ filter. For example, with the IA-64 filter,
958 with LZMA2 (2^4=16). The x86 filter is an exception; it's usually good to
959 stick to LZMA2's default four-byte alignment when compressing x86 executables.
961 All BCJ filters support the same
968 that is used when converting between relative and absolute addresses.
971 must be a multiple of the alignment of the filter (see the table above).
972 The default is zero. In practice, the default is good; specifying
975 is almost never useful.
977 Specifying a non-zero start
979 is probably useful only if the executable has multiple sections, and there
980 are many cross-section jumps or calls. Applying a BCJ filter separately for
981 each section with proper start offset and then compressing the result as
982 a single chunk may give some improvement in compression ratio compared
983 to applying the BCJ filter with the default
985 for the whole executable.
988 \fB\-\-delta\fR[\fB=\fIoptions\fR]
989 Add Delta filter to the filter chain. The Delta filter
990 can be used only as non-last filter in the filter chain.
992 Currently only simple byte-wise delta calculation is supported. It can
993 be useful when compressing e.g. uncompressed bitmap images or uncompressed
994 PCM audio. However, special purpose algorithms may give significantly better
995 results than Delta + LZMA2. This is true especially with audio, which
996 compresses faster and better e.g. with FLAC.
1005 of the delta calculation as bytes.
1007 must be 1\-256. The default is 1.
1011 and eight-byte input A1 B1 A2 B3 A3 B5 A4 B7, the output will be
1012 A1 B1 01 02 01 02 01 02.
1016 .BR \-q ", " \-\-quiet
1017 Suppress warnings and notices. Specify this twice to suppress errors too.
1018 This option has no effect on the exit status. That is, even if a warning
1019 was suppressed, the exit status to indicate a warning is still used.
1021 .BR \-v ", " \-\-verbose
1022 Be verbose. If standard error is connected to a terminal,
1024 will display a progress indicator.
1027 twice will give even more verbose output (useful mostly for debugging).
1029 The progress indicator shows the following information:
1032 Completion percentage is shown if the size of the input file is known.
1033 That is, percentage cannot be shown in pipes.
1035 Amount of compressed data produced (compressing) or consumed (decompressing).
1037 Amount of uncompressed data consumed (compressing) or produced
1040 Compression ratio, which is calculated by dividing the amount of
1041 compressed data processed so far by the amount of uncompressed data
1044 Compression or decompression speed. This is measured as the amount of
1045 uncompressed data consumed (compression) or produced (decompression)
1046 per second. It is shown once a few seconds have passed since
1048 started processing the file.
1050 Elapsed time or estimated time remaining.
1051 Elapsed time is displayed in the format M:SS or H:MM:SS.
1052 The estimated remaining time is displayed in a less precise format
1053 which never has colons, for example, 2 min 30 s. The estimate can
1054 be shown only when the size of the input file is known and a couple of
1055 seconds have already passed since
1057 started processing the file.
1060 When standard error is not a terminal,
1064 print the filename, compressed size, uncompressed size, compression ratio,
1065 speed, and elapsed time on a single line to standard error after
1066 compressing or decompressing the file. If operating took at least a few
1067 seconds, also the speed and elapsed time are printed. If the operation
1068 didn't finish, for example due to user interruption, also the completion
1069 percentage is printed if the size of the input file is known.
1071 .BR \-Q ", " \-\-no\-warn
1072 Don't set the exit status to
1074 even if a condition worth a warning was detected. This option doesn't affect
1075 the verbosity level, thus both
1079 have to be used to not display warnings and to not alter the exit status.
1082 Print messages in a machine-parsable format. This is intended to ease
1083 writing frontends that want to use
1085 instead of liblzma, which may be the case with various scripts. The output
1086 with this option enabled is meant to be stable across
1090 is implemented only for
1094 but the idea is to make it usable for actual compression
1095 and decompression too.
1098 Display the current memory usage limit in human-readable format on
1099 a single line, and exit successfully. To see how much RAM
1101 thinks your system has, use
1102 .BR "\-\-memory=100% \-\-info\-memory" .
1103 To get machine-parsable output
1104 (memory usage limit as bytes without thousand separators), specify
1107 .BR \-\-info-memory .
1109 .BR \-h ", " \-\-help
1110 Display a help message describing the most commonly used options,
1111 and exit successfully.
1113 .BR \-H ", " \-\-long\-help
1114 Display a help message describing all features of
1116 and exit successfully
1118 .BR \-V ", " \-\-version
1119 Display the version number of
1121 and liblzma in human readable format. To get machine-parsable output, specify
1134 Something worth a warning occurred, but no actual errors occurred.
1136 Notices (not warnings or errors) printed on standard error don't affect
1141 A space-separated list of options is parsed from
1143 before parsing the options given on the command line. Note that only
1144 options are parsed from
1146 all non-options are silently ignored. Parsing is done with
1148 which is used also for the command line arguments.
1149 .SH "LZMA UTILS COMPATIBILITY"
1150 The command line syntax of
1152 is practically a superset of
1157 as found from LZMA Utils 4.32.x. In most cases, it is possible to replace
1158 LZMA Utils with XZ Utils without breaking existing scripts. There are some
1159 incompatibilities though, which may sometimes cause problems.
1160 .SS "Compression preset levels"
1161 The numbering of the compression level presets is not identical in
1164 The most important difference is how dictionary sizes are mapped to different
1165 presets. Dictionary size is roughly equal to the decompressor memory usage.
1184 The dictionary size differences affect the compressor memory usage too,
1185 but there are some other differences between LZMA Utils and XZ Utils, which
1186 make the difference even bigger:
1192 Level;xz;LZMA Utils 4.32.x
1205 The default preset level in LZMA Utils is
1207 while in XZ Utils it is
1209 so both use 8 MiB dictionary by default.
1210 .SS "Streamed vs. non-streamed .lzma files"
1211 Uncompressed size of the file can be stored in the
1213 header. LZMA Utils does that when compressing regular files.
1214 The alternative is to mark that uncompressed size is unknown and
1215 use end of payload marker to indicate where the decompressor should stop.
1216 LZMA Utils uses this method when uncompressed size isn't known, which is
1217 the case for example in pipes.
1220 supports decompressing
1222 files with or without end of payload marker, but all
1226 will use end of payload marker and have uncompressed size marked as unknown
1229 header. This may be a problem in some (uncommon) situations. For example, a
1231 decompressor in an embedded device might work only with files that have known
1232 uncompressed size. If you hit this problem, you need to use LZMA Utils or
1235 files with known uncompressed size.
1236 .SS "Unsupported .lzma files"
1243 values up to 4. LZMA Utils can decompress files with any
1247 but always creates files with
1251 Creating files with other
1259 The implementation of the LZMA1 filter in liblzma requires
1264 must not exceed 4. Thus,
1266 files which exceed this limitation, cannot be decompressed with
1269 LZMA Utils creates only
1271 files which have dictionary size of
1273 (a power of 2), but accepts files with any dictionary size.
1274 liblzma accepts only
1276 files which have dictionary size of
1279 .RI "2^" n " + 2^(" n "\-1)."
1280 This is to decrease false positives when detecting
1284 These limitations shouldn't be a problem in practice, since practically all
1286 files have been compressed with settings that liblzma will accept.
1287 .SS "Trailing garbage"
1288 When decompressing, LZMA Utils silently ignore everything after the first
1290 stream. In most situations, this is a bug. This also means that LZMA Utils
1291 don't support decompressing concatenated
1295 If there is data left after the first
1299 considers the file to be corrupt. This may break obscure scripts which have
1300 assumed that trailing garbage is ignored.
1302 .SS Compressed output may vary
1303 The exact compressed output produced from the same uncompressed input file
1304 may vary between XZ Utils versions even if compression options are identical.
1305 This is because the encoder can be improved (faster or better compression)
1306 without affecting the file format. The output can vary even between different
1307 builds of the same XZ Utils version, if different build options are used.
1309 The above means that implementing
1313 files is not going to happen without freezing a part of the encoder
1314 implementation, which can then be used with
1316 .SS Embedded .xz decompressors
1319 decompressor implementations like XZ Embedded don't necessarily support files
1326 Since the default is \fB\-\-check=\fIcrc64\fR, you must use
1330 when creating files for embedded systems.
1332 Outside embedded systems, all
1334 format decompressors support all the
1336 types, or at least are able to decompress the file without verifying the
1337 integrity check if the particular
1341 XZ Embedded supports BCJ filters, but only with the default start offset.
1347 XZ Utils: <http://tukaani.org/xz/>
1349 XZ Embedded: <http://tukaani.org/xz/embedded.html>
1351 LZMA SDK: <http://7-zip.org/sdk.html>