xz.1		xz.1
	XZ(1) XZ Utils XZ (1)		XZ(1) XZ Utils XZ (1)
	NAME		NAME
	xz, unxz, xzcat, lzma, unlzma, lzcat - Compress or decompress .xz and		xz, unxz, xzcat, lzma, unlzma, lzcat - Compress or decompress .xz and
	.lzma files		.lzma files
	SYNOPSIS		SYNOPSIS
	xz [option]... [file]...		xz [option...] [file...]
			COMMAND ALIASES
	unxz is equivalent to xz --decompress.		unxz is equivalent to xz --decompress.
	xzcat is equivalent to xz --decompress --stdout.		xzcat is equivalent to xz --decompress --stdout.
	lzma is equivalent to xz --format=lzma.		lzma is equivalent to xz --format=lzma.
	unlzma is equivalent to xz --format=lzma --decompress.		unlzma is equivalent to xz --format=lzma --decompress.
	lzcat is equivalent to xz --format=lzma --decompress --stdout.		lzcat is equivalent to xz --format=lzma --decompress --stdout.
	When writing scripts that need to decompress files, it is recommen ded		When writing scripts that need to decompress files, it is recommen ded
	to always use the name xz with appropriate arguments (xz -d or xz - dc)		to always use the name xz with appropriate arguments (xz -d or xz - dc)
	instead of the names unxz and xzcat.		instead of the names unxz and xzcat.
	skipping to change at line 317		skipping to change at line 318
	is the default, since it is slightly better than CRC32 at		is the default, since it is slightly better than CRC32 at
	detecting damaged files and the speed difference is n eg‐		detecting damaged files and the speed difference is n eg‐
	ligible.		ligible.
	sha256 Calculate SHA-256. This is somewhat slower than CR C32		sha256 Calculate SHA-256. This is somewhat slower than CR C32
	and CRC64.		and CRC64.
	Integrity of the .xz headers is always verified with CRC32. It		Integrity of the .xz headers is always verified with CRC32. It
	is not possible to change or disable it.		is not possible to change or disable it.
			--ignore-check
			Don't verify the integrity check of the compressed data w
			hen
			decompressing. The CRC32 values in the .xz headers will st
			ill
			be verified normally.
			Do not use this option unless you know what you are doing. P
			os‐
			sible reasons to use this option:
			· Trying to recover data from a corrupt .xz file.
			· Speeding up decompression. This matters mostly with SHA-
			256
			or with files that have compressed extremely well. It's r
			ec‐
			ommended to not use this option for this purpose unless
			the
			file integrity is verified externally in some other way.
	-0 ... -9		-0 ... -9
	Select a compression preset level. The default is -6. If m ul‐		Select a compression preset level. The default is -6. If m ul‐
	tiple preset levels are specified, the last one takes effe ct.		tiple preset levels are specified, the last one takes effe ct.
	If a custom filter chain was already specified, setting a c om‐		If a custom filter chain was already specified, setting a c om‐
	pression preset level clears the custom filter chain.		pression preset level clears the custom filter chain.
	The differences between the presets are more significant t han		The differences between the presets are more significant t han
	with gzip(1) and bzip2(1). The selected compression setti ngs		with gzip(1) and bzip2(1). The selected compression setti ngs
	determine the memory requirements of the decompressor, t hus		determine the memory requirements of the decompressor, t hus
	using a too high preset level might make it painful to dec om‐		using a too high preset level might make it painful to dec om‐
	skipping to change at line 396		skipping to change at line 412
	· CompMem contains the compressor memory requirements in the		· CompMem contains the compressor memory requirements in the
	single-threaded mode. It may vary slightly between xz v er‐		single-threaded mode. It may vary slightly between xz v er‐
	sions. Memory requirements of some of the future mul ti‐		sions. Memory requirements of some of the future mul ti‐
	threaded modes may be dramatically higher than that of the		threaded modes may be dramatically higher than that of the
	single-threaded mode.		single-threaded mode.
	· DecMem contains the decompressor memory requirements. T hat		· DecMem contains the decompressor memory requirements. T hat
	is, the compression settings determine the memory requi re‐		is, the compression settings determine the memory requi re‐
	ments of the decompressor. The exact decompressor mem ory		ments of the decompressor. The exact decompressor mem ory
	usage is slighly more than the LZMA2 dictionary size, but		usage is slightly more than the LZMA2 dictionary size,
	the		but
	values in the table have been rounded up to the next f		the values in the table have been rounded up to the next f
	ull		ull
	MiB.		MiB.
	-e, --extreme		-e, --extreme
	Use a slower variant of the selected compression preset le vel		Use a slower variant of the selected compression preset le vel
	(-0 ... -9) to hopefully get a little bit better compress ion		(-0 ... -9) to hopefully get a little bit better compress ion
	ratio, but with bad luck this can also make it worse. Dec om‐		ratio, but with bad luck this can also make it worse. Dec om‐
	pressor memory usage is not affected, but compressor mem ory		pressor memory usage is not affected, but compressor mem ory
	usage increases a little at preset levels -0 ... -3.		usage increases a little at preset levels -0 ... -3.
	Since there are two presets with dictionary sizes 4 MiB and		Since there are two presets with dictionary sizes 4 MiB and
	skipping to change at line 436		skipping to change at line 452
	-6, -5e, and -6e.		-6, -5e, and -6e.
	--fast		--fast
	--best These are somewhat misleading aliases for -0 and -9, resp ec‐		--best These are somewhat misleading aliases for -0 and -9, resp ec‐
	tively. These are provided only for backwards compatibil ity		tively. These are provided only for backwards compatibil ity
	with LZMA Utils. Avoid using these options.		with LZMA Utils. Avoid using these options.
	--block-size=size		--block-size=size
	When compressing to the .xz format, split the input data i nto		When compressing to the .xz format, split the input data i nto
	blocks of size bytes. The blocks are compressed independen tly		blocks of size bytes. The blocks are compressed independen tly
	from each other.		from each other, which helps with multi-threading and makes l
			im‐
			ited random-access decompression possible. This option is ty
			pi‐
			cally used to override the default block size in multi-threa
			ded
			mode, but this option can be used in single-threaded mode too
			.
			In multi-threaded mode about three times size bytes will
			be
			allocated in each thread for buffering input and output.
			The
			default size is three times the LZMA2 dictionary size or 1 M
			iB,
			whichever is more. Typically a good value is 2-4 times the s
			ize
			of the LZMA2 dictionary or at least 1 MiB. Using size less t
			han
			the LZMA2 dictionary size is waste of RAM because then the LZ
			MA2
			dictionary buffer will never get fully used. The sizes of
			the
			blocks are stored in the block headers, which a future vers
			ion
			of xz will use for multi-threaded decompression.
			In single-threaded mode no block splitting is done by defau
			lt.
			Setting this option doesn't affect memory usage. No size inf
			or‐
			mation is stored in block headers, thus files created in sing
			le-
			threaded mode won't be identical to files created in mul
			ti-
			threaded mode. The lack of size information also means tha
			t a
			future version of xz won't be able decompress the files
			in
			multi-threaded mode.
	--block-list=sizes		--block-list=sizes
	When compressing to the .xz format, start a new block after the		When compressing to the .xz format, start a new block after the
	given intervals of uncompressed data.		given intervals of uncompressed data.
	The uncompressed sizes of the blocks are specified as a com		The uncompressed sizes of the blocks are specified as a com
	ma-		ma-
	separated list. Omitting a size (two or more consecutive c		separated list. Omitting a size (two or more consecutive c
	om‐		om‐
	mas) is a shorthand to use the size of the previous block.		mas) is a shorthand to use the size of the previous block.
	A
	special value of 0 may be used as the last value to indic		If the input file is bigger than the sum of sizes, the l
	ate		ast
	that the rest of the file should be encoded as a single block		value in sizes is repeated until the end of the file. A spec
	.		ial
			value of 0 may be used as the last value to indicate that
	Currently this option is badly broken if used together w		the
	ith		rest of the file should be encoded as a single block.
	--block-size or with multithreading.
			If one specifies sizes that exceed the encoder's block s
			ize
			(either the default value in threaded mode or the value spe
			ci‐
			fied with --block-size=size), the encoder will create additio
			nal
			blocks while keeping the boundaries specified in sizes.
			For
			example, if one specifies --block-size=10
			MiB
			--block-list=5MiB,10MiB,8MiB,12MiB,24MiB and the input file
			is
			80 MiB, one will get 11 blocks: 5, 10, 8, 10, 2, 10, 10, 4,
			10,
			10, and 1 MiB.
			In multi-threaded mode the sizes of the blocks are stored in
			the
			block headers. This isn't done in single-threaded mode, so
			the
			encoded output won't be identical to that of the multi-threa
			ded
			mode.
			--flush-timeout=timeout
			When compressing, if more than timeout milliseconds (a posit
			ive
			integer) has passed since the previous flush and reading m
			ore
			input would block, all the pending input data is flushed f
			rom
			the encoder and made available in the output stream. This
			can
			be useful if xz is used to compress data that is streamed ove
			r a
			network. Small timeout values make the data available at
			the
			receiving end with a small delay, but large timeout values g
			ive
			better compression ratio.
			This feature is disabled by default. If this option is spe
			ci‐
			fied more than once, the last one takes effect. The spec
			ial
			timeout value of 0 can be used to explicitly disable this f
			ea‐
			ture.
			This feature is not available on non-POSIX systems.
			This feature is still experimental. Currently xz is unsuita
			ble
			for decompressing the stream in real time due to how xz d
			oes
			buffering.
	--memlimit-compress=limit		--memlimit-compress=limit
	Set a memory usage limit for compression. If this option is		Set a memory usage limit for compression. If this option is
	specified multiple times, the last one takes effect.		specified multiple times, the last one takes effect.
	If the compression settings exceed the limit, xz will adjust the		If the compression settings exceed the limit, xz will adjust the
	settings downwards so that the limit is no longer exceeded		settings downwards so that the limit is no longer exceeded
	and		and
	display a notice that automatic adjustment was done. S		display a notice that automatic adjustment was done. S
	uch		uch
	adjustments are not made when compressing with --format=raw		adjustments are not made when compressing with --format=raw
	or		or
	if --no-adjust has been specified. In those cases, an error		if --no-adjust has been specified. In those cases, an error
	is		is
	displayed and xz will exit with exit status 1.		displayed and xz will exit with exit status 1.
	The limit can be specified in multiple ways:		The limit can be specified in multiple ways:
	· The limit can be an absolute value in bytes. Using an in		· The limit can be an absolute value in bytes. Using an in
	te‐		te‐
	ger suffix like MiB can be useful. Example: --memlimit-c		ger suffix like MiB can be useful. Example: --memlimit-c
	om‐		om‐
	press=80MiB		press=80MiB
	· The limit can be specified as a percentage of total physi cal		· The limit can be specified as a percentage of total physi cal
	memory (RAM). This can be useful especially when setting the		memory (RAM). This can be useful especially when setting the
	XZ_DEFAULTS environment variable in a shell initializat		XZ_DEFAULTS environment variable in a shell initializat
	ion		ion
	script that is shared between different computers. That		script that is shared between different computers. That
	way		way
	the limit is automatically bigger on systems with more m		the limit is automatically bigger on systems with more m
	em‐		em‐
	ory. Example: --memlimit-compress=70%		ory. Example: --memlimit-compress=70%
	· The limit can be reset back to its default value by sett		· The limit can be reset back to its default value by sett
	ing		ing
	it to 0. This is currently equivalent to setting the li		it to 0. This is currently equivalent to setting the li
	mit		mit
	to max (no memory usage limit). Once multithreading supp		to max (no memory usage limit). Once multithreading supp
	ort		ort
	has been implemented, there may be a difference between 0 and		has been implemented, there may be a difference between 0 and
	max for the multithreaded case, so it is recommended to us e 0		max for the multithreaded case, so it is recommended to us e 0
	instead of max until the details have been decided.		instead of max until the details have been decided.
	See also the section Memory usage.		See also the section Memory usage.
	--memlimit-decompress=limit		--memlimit-decompress=limit
	Set a memory usage limit for decompression. This also affe		Set a memory usage limit for decompression. This also affe
	cts		cts
	the --list mode. If the operation is not possible with		the --list mode. If the operation is not possible with
	out		out
	exceeding the limit, xz will display an error and decompress		exceeding the limit, xz will display an error and decompress
	ing		ing
	the file will fail. See --memlimit-compress=limit for possi		the file will fail. See --memlimit-compress=limit for possi
	ble		ble
	ways to specify the limit.		ways to specify the limit.
	-M limit, --memlimit=limit, (old alias --memory=limit)		-M limit, --memlimit=limit, --memory=limit
	This is equivalent to specifying --memlimit-compress=li		This is equivalent to specifying --memlimit-compress=li
	mit		mit
	--memlimit-decompress=limit.		--memlimit-decompress=limit.
	--no-adjust		--no-adjust
	Display an error and exit if the compression settings exceed the		Display an error and exit if the compression settings exceed the
	memory usage limit. The default is to adjust the settings do wn‐		memory usage limit. The default is to adjust the settings do wn‐
	wards so that the memory usage limit is not exceeded. Automa tic		wards so that the memory usage limit is not exceeded. Automa tic
	adjusting is always disabled when creating raw streams (--f or‐		adjusting is always disabled when creating raw streams (--f or‐
	mat=raw).		mat=raw).
	-T threads, --threads=threads		-T threads, --threads=threads
	Specify the number of worker threads to use. Setting threads to		Specify the number of worker threads to use. Setting threads to
	a special value 0 makes xz use as many threads as there are		a special value 0 makes xz use as many threads as there are
	CPU		CPU
	cores on the system. The actual number of threads can be l		cores on the system. The actual number of threads can be l
	ess		ess
	than threads if the input file is not big enough for thread		than threads if the input file is not big enough for thread
	ing		ing
	with the given settings or if using more threads would exc		with the given settings or if using more threads would exc
	eed		eed
	the memory usage limit.		the memory usage limit.
	Currently the only threading method is to split the input i		Currently the only threading method is to split the input i
	nto		nto
	blocks and compress them independently from each other.		blocks and compress them independently from each other.
	The		The
	default block size depends on the compression level and can		default block size depends on the compression level and can
	be		be
	overriden with the --block-size=size option.		overriden with the --block-size=size option.
	It is possible that the details of this option change before
	the
	next stable XZ Utils release. This may include the meaning
	of
	the special value 0.
	Custom compressor filter chains		Custom compressor filter chains
	A custom filter chain allows specifying the compression settings		A custom filter chain allows specifying the compression settings
	in		in
	detail instead of relying on the settings associated to the preset l		detail instead of relying on the settings associated to the prese
	ev‐		ts.
	els. When a custom filter chain is specified, the compression pre		When a custom filter chain is specified, preset options (-0 ... -9
	set		and
	level options (-0 ... -9 and --extreme) are silently ignored.		--extreme) earlier on the command line are forgotten. If a pre
			set
			option is specified after one or more custom filter chain options,
			the
			new preset takes effect and the custom filter chain options specif
			ied
			earlier are forgotten.
	A filter chain is comparable to piping on the command line. When c om‐		A filter chain is comparable to piping on the command line. When c om‐
	pressing, the uncompressed input goes to the first filter, whose out put		pressing, the uncompressed input goes to the first filter, whose out put
	goes to the next filter (if any). The output of the last filter g ets		goes to the next filter (if any). The output of the last filter g ets
	written to the compressed file. The maximum number of filters in the		written to the compressed file. The maximum number of filters in the
	chain is four, but typically a filter chain has only one or two f il‐		chain is four, but typically a filter chain has only one or two f il‐
	ters.		ters.
	Many filters have limitations on where they can be in the filter cha in:		Many filters have limitations on where they can be in the filter cha in:
	some filters can work only as the last filter in the chain, some o nly		some filters can work only as the last filter in the chain, some o nly
	skipping to change at line 548		skipping to change at line 619
	A custom filter chain is specified by using one or more filter opti ons		A custom filter chain is specified by using one or more filter opti ons
	in the order they are wanted in the filter chain. That is, the or der		in the order they are wanted in the filter chain. That is, the or der
	of filter options is significant! When decoding raw streams (--f or‐		of filter options is significant! When decoding raw streams (--f or‐
	mat=raw), the filter chain is specified in the same order as it was		mat=raw), the filter chain is specified in the same order as it was
	specified when compressing.		specified when compressing.
	Filters take filter-specific options as a comma-separated list. Ex tra		Filters take filter-specific options as a comma-separated list. Ex tra
	commas in options are ignored. Every option has a default value, so		commas in options are ignored. Every option has a default value, so
	you need to specify only those you want to change.		you need to specify only those you want to change.
			To see the whole filter chain and options, use xz -vv (that is,
			use
			--verbose twice). This works also for viewing the filter chain opti
			ons
			used by presets.
	--lzma1[=options]		--lzma1[=options]
	--lzma2[=options]		--lzma2[=options]
	Add LZMA1 or LZMA2 filter to the filter chain. These filt ers		Add LZMA1 or LZMA2 filter to the filter chain. These filt ers
	can be used only as the last filter in the chain.		can be used only as the last filter in the chain.
	LZMA1 is a legacy filter, which is supported almost solely due		LZMA1 is a legacy filter, which is supported almost solely due
	to the legacy .lzma file format, which supports only LZM A1.		to the legacy .lzma file format, which supports only LZM A1.
	LZMA2 is an updated version of LZMA1 to fix some practi cal		LZMA2 is an updated version of LZMA1 to fix some practi cal
	issues of LZMA1. The .xz format uses LZMA2 and doesn't supp ort		issues of LZMA1. The .xz format uses LZMA2 and doesn't supp ort
	LZMA1 at all. Compression speed and ratios of LZMA1 and LZ MA2		LZMA1 at all. Compression speed and ratios of LZMA1 and LZ MA2
	are practically the same.		are practically the same.
	LZMA1 and LZMA2 share the same set of options:		LZMA1 and LZMA2 share the same set of options:
	preset=preset		preset=preset
	Reset all LZMA1 or LZMA2 options to preset. Preset c on‐		Reset all LZMA1 or LZMA2 options to preset. Preset c on‐
	sist of an integer, which may be followed by single-l et‐		sist of an integer, which may be followed by single-l et‐
	ter preset modifiers. The integer can be from 0 to 9,		ter preset modifiers. The integer can be from 0 to 9,
	matching the command line options -0 ... -9. The o nly		matching the command line options -0 ... -9. The o nly
	supported modifier is currently e, which matc hes		supported modifier is currently e, which matc hes
	--extreme. The default preset is 6, from which		--extreme. If no preset is specified, the default val
	the		ues
	default values for the rest of the LZMA1 or LZMA2 opti		of LZMA1 or LZMA2 options are taken from the preset 6.
	ons
	are taken.
	dict=size		dict=size
	Dictionary (history buffer) size indicates how many by tes		Dictionary (history buffer) size indicates how many by tes
	of the recently processed uncompressed data is kept		of the recently processed uncompressed data is kept
	in		in
	memory. The algorithm tries to find repeating b		memory. The algorithm tries to find repeating b
	yte		yte
	sequences (matches) in the uncompressed data, and repl ace		sequences (matches) in the uncompressed data, and repl ace
	them with references to the data currently in the dict io‐		them with references to the data currently in the dict io‐
	nary. The bigger the dictionary, the higher is the		nary. The bigger the dictionary, the higher is the
	chance to find a match. Thus, increasing dictionary s ize		chance to find a match. Thus, increasing dictionary s ize
	usually improves compression ratio, but a dictionary b ig‐		usually improves compression ratio, but a dictionary b ig‐
	ger than the uncompressed file is waste of memory.		ger than the uncompressed file is waste of memory.
	Typical dictionary size is from 64 KiB to 64 MiB.		Typical dictionary size is from 64 KiB to 64 MiB.
	The		The
	minimum is 4 KiB. The maximum for compression is c		minimum is 4 KiB. The maximum for compression is c
	ur‐		ur‐
	rently 1.5 GiB (1536 MiB). The decompressor already s up‐		rently 1.5 GiB (1536 MiB). The decompressor already s up‐
	ports dictionaries up to one byte less than 4 GiB, wh ich		ports dictionaries up to one byte less than 4 GiB, wh ich
	is the maximum for the LZMA1 and LZMA2 stream formats.		is the maximum for the LZMA1 and LZMA2 stream formats.
	Dictionary size and match finder (mf) together determ ine		Dictionary size and match finder (mf) together determ ine
	the memory usage of the LZMA1 or LZMA2 encoder. The s ame		the memory usage of the LZMA1 or LZMA2 encoder. The s ame
	(or bigger) dictionary size is required for decompress ing		(or bigger) dictionary size is required for decompress ing
	that was used when compressing, thus the memory usage		that was used when compressing, thus the memory usage
	of		of
	the decoder is determined by the dictionary size u		the decoder is determined by the dictionary size u
	sed		sed
	when compressing. The .xz headers store the diction		when compressing. The .xz headers store the diction
	ary		ary
	size either as 2^n or 2^n + 2^(n-1), so these sizes		size either as 2^n or 2^n + 2^(n-1), so these sizes
	are		are
	somewhat preferred for compression. Other sizes will get		somewhat preferred for compression. Other sizes will get
	rounded up when stored in the .xz headers.		rounded up when stored in the .xz headers.
	lc=lc Specify the number of literal context bits. The mini		lc=lc Specify the number of literal context bits. The mini
	mum		mum
	is 0 and the maximum is 4; the default is 3. In ad		is 0 and the maximum is 4; the default is 3. In ad
	di‐		di‐
	tion, the sum of lc and lp must not exceed 4.		tion, the sum of lc and lp must not exceed 4.
	All bytes that cannot be encoded as matches are enco		All bytes that cannot be encoded as matches are enco
	ded		ded
	as literals. That is, literals are simply 8-bit by		as literals. That is, literals are simply 8-bit by
	tes		tes
	that are encoded one at a time.		that are encoded one at a time.
	The literal coding makes an assumption that the high		The literal coding makes an assumption that the high
	est		est
	lc bits of the previous uncompressed byte correlate w		lc bits of the previous uncompressed byte correlate w
	ith		ith
	the next byte. E.g. in typical English text, an upp		the next byte. E.g. in typical English text, an upp
	er-		er-
	case letter is often followed by a lower-case letter, and		case letter is often followed by a lower-case letter, and
	a lower-case letter is usually followed by another low er-		a lower-case letter is usually followed by another low er-
	case letter. In the US-ASCII character set, the high		case letter. In the US-ASCII character set, the high
	est		est
	three bits are 010 for upper-case letters and 011		three bits are 010 for upper-case letters and 011
	for		for
	lower-case letters. When lc is at least 3, the lite		lower-case letters. When lc is at least 3, the lite
	ral		ral
	coding can take advantage of this property in the unc		coding can take advantage of this property in the unc
	om‐		om‐
	pressed data.		pressed data.
	The default value (3) is usually good. If you want ma xi‐		The default value (3) is usually good. If you want ma xi‐
	mum compression, test lc=4. Sometimes it helps a litt le,		mum compression, test lc=4. Sometimes it helps a litt le,
	and sometimes it makes compression worse. If it makes it		and sometimes it makes compression worse. If it makes it
	worse, test e.g. lc=2 too.		worse, test e.g. lc=2 too.
	lp=lp Specify the number of literal position bits. The mini mum		lp=lp Specify the number of literal position bits. The mini mum
	is 0 and the maximum is 4; the default is 0.		is 0 and the maximum is 4; the default is 0.
	Lp affects what kind of alignment in the uncompres sed		Lp affects what kind of alignment in the uncompres sed
	data is assumed when encoding literals. See pb below for		data is assumed when encoding literals. See pb below for
	more information about alignment.		more information about alignment.
	pb=pb Specify the number of position bits. The minimum i s 0		pb=pb Specify the number of position bits. The minimum is 0
	and the maximum is 4; the default is 2.		and the maximum is 4; the default is 2.
	Pb affects what kind of alignment in the uncompres		Pb affects what kind of alignment in the uncompres
	sed		sed
	data is assumed in general. The default means four-b		data is assumed in general. The default means four-b
	yte		yte
	alignment (2^pb=2^2=4), which is often a good choice w hen		alignment (2^pb=2^2=4), which is often a good choice w hen
	there's no better guess.		there's no better guess.
	When the aligment is known, setting pb accordingly may		When the aligment is known, setting pb accordingly may
	reduce the file size a little. E.g. with text files h av‐		reduce the file size a little. E.g. with text files h av‐
	ing one-byte alignment (US-ASCII, ISO-8859-*, UTF- 8),		ing one-byte alignment (US-ASCII, ISO-8859-*, UTF- 8),
	setting pb=0 can improve compression slightly. For		setting pb=0 can improve compression slightly. For
	UTF-16 text, pb=1 is a good choice. If the alignment		UTF-16 text, pb=1 is a good choice. If the alignment
	is		is
	an odd number like 3 bytes, pb=0 might be the b		an odd number like 3 bytes, pb=0 might be the b
	est		est
	choice.		choice.
	Even though the assumed alignment can be adjusted with pb		Even though the assumed alignment can be adjusted with pb
	and lp, LZMA1 and LZMA2 still slightly favor 16-b		and lp, LZMA1 and LZMA2 still slightly favor 16-b
	yte		yte
	alignment. It might be worth taking into account w		alignment. It might be worth taking into account w
	hen		hen
	designing file formats that are likely to be often c		designing file formats that are likely to be often c
	om‐		om‐
	pressed with LZMA1 or LZMA2.		pressed with LZMA1 or LZMA2.
	mf=mf Match finder has a major effect on encoder speed, mem		mf=mf Match finder has a major effect on encoder speed, mem
	ory		ory
	usage, and compression ratio. Usually Hash Chain ma		usage, and compression ratio. Usually Hash Chain ma
	tch		tch
	finders are faster than Binary Tree match finders.		finders are faster than Binary Tree match finders.
	The		The
	default depends on the preset: 0 uses hc3, 1-3 use h		default depends on the preset: 0 uses hc3, 1-3 use h
	c4,		c4,
	and the rest use bt4.		and the rest use bt4.
	The following match finders are supported. The mem		The following match finders are supported. The mem
	ory		ory
	usage formulas below are rough approximations, which		usage formulas below are rough approximations, which
	are		are
	closest to the reality when dict is a power of two.		closest to the reality when dict is a power of two.
	hc3 Hash Chain with 2- and 3-byte hashing		hc3 Hash Chain with 2- and 3-byte hashing
	Minimum value for nice: 3		Minimum value for nice: 3
	Memory usage:		Memory usage:
	dict * 7.5 (if dict <= 16 MiB);		dict * 7.5 (if dict <= 16 MiB);
	dict * 5.5 + 64 MiB (if dict > 16 MiB)		dict * 5.5 + 64 MiB (if dict > 16 MiB)
	hc4 Hash Chain with 2-, 3-, and 4-byte hashing		hc4 Hash Chain with 2-, 3-, and 4-byte hashing
	Minimum value for nice: 4		Minimum value for nice: 4
	skipping to change at line 692		skipping to change at line 766
	dict * 9.5 + 64 MiB (if dict > 16 MiB)		dict * 9.5 + 64 MiB (if dict > 16 MiB)
	bt4 Binary Tree with 2-, 3-, and 4-byte hashing		bt4 Binary Tree with 2-, 3-, and 4-byte hashing
	Minimum value for nice: 4		Minimum value for nice: 4
	Memory usage:		Memory usage:
	dict * 11.5 (if dict <= 32 MiB);		dict * 11.5 (if dict <= 32 MiB);
	dict * 10.5 (if dict > 32 MiB)		dict * 10.5 (if dict > 32 MiB)
	mode=mode		mode=mode
	Compression mode specifies the method to analyze the d ata		Compression mode specifies the method to analyze the d ata
	produced by the match finder. Supported modes are f ast		produced by the match finder. Supported modes are f ast
	and normal. The default is fast for presets 0-3 and n or‐		and normal. The default is fast for presets 0-3 and n or‐
	mal for presets 4-9.		mal for presets 4-9.
	Usually fast is used with Hash Chain match finders and		Usually fast is used with Hash Chain match finders and
	normal with Binary Tree match finders. This is also w hat		normal with Binary Tree match finders. This is also w hat
	the presets do.		the presets do.
	nice=nice		nice=nice
	Specify what is considered to be a nice length for a		Specify what is considered to be a nice length fo r a
	match. Once a match of at least nice bytes is found, the		match. Once a match of at least nice bytes is found, the
	algorithm stops looking for possibly better matches.		algorithm stops looking for possibly better matches.
	Nice can be 2-273 bytes. Higher values tend to give b et‐		Nice can be 2-273 bytes. Higher values tend to give b et‐
	ter compression ratio at the expense of speed. The		ter compression ratio at the expense of speed. The
	default depends on the preset.		default depends on the preset.
	depth=depth		depth=depth
	Specify the maximum search depth in the match find		Specify the maximum search depth in the match find
	er.		er.
	The default is the special value of 0, which makes		The default is the special value of 0, which makes
	the		the
	compressor determine a reasonable depth from mf and ni ce.		compressor determine a reasonable depth from mf and ni ce.
	Reasonable depth for Hash Chains is 4-100 and 16-1000 for		Reasonable depth for Hash Chains is 4-100 and 16-1000 for
	Binary Trees. Using very high values for depth can m		Binary Trees. Using very high values for depth can m
	ake		ake
	the encoder extremely slow with some files. Avoid s		the encoder extremely slow with some files. Avoid s
	et‐		et‐
	ting the depth over 1000 unless you are prepared		ting the depth over 1000 unless you are prepared
	to		to
	interrupt the compression in case it is taking far		interrupt the compression in case it is taking far
	too		too
	long.		long.
	When decoding raw streams (--format=raw), LZMA2 needs only the		When decoding raw streams (--format=raw), LZMA2 needs only the
	dictionary size. LZMA1 needs also lc, lp, and pb.		dictionary size. LZMA1 needs also lc, lp, and pb.
	--x86[=options]		--x86[=options]
	--powerpc[=options]		--powerpc[=options]
	--ia64[=options]		--ia64[=options]
	--arm[=options]		--arm[=options]
	--armthumb[=options]		--armthumb[=options]
	--sparc[=options]		--sparc[=options]
	Add a branch/call/jump (BCJ) filter to the filter chain. Th		Add a branch/call/jump (BCJ) filter to the filter chain. Th
	ese		ese
	filters can be used only as a non-last filter in the fil		filters can be used only as a non-last filter in the fil
	ter		ter
	chain.		chain.
	A BCJ filter converts relative addresses in the machine code		A BCJ filter converts relative addresses in the machine code
	to		to
	their absolute counterparts. This doesn't change the size		their absolute counterparts. This doesn't change the size
	of		of
	the data, but it increases redundancy, which can help LZMA2		the data, but it increases redundancy, which can help LZMA2
	to		to
	produce 0-15 % smaller .xz file. The BCJ filters are alw		produce 0-15 % smaller .xz file. The BCJ filters are alw
	ays		ays
	reversible, so using a BCJ filter for wrong type of data does n't		reversible, so using a BCJ filter for wrong type of data does n't
	cause any data loss, although it may make the compression ra tio		cause any data loss, although it may make the compression ra tio
	slightly worse.		slightly worse.
	It is fine to apply a BCJ filter on a whole executable; ther		It is fine to apply a BCJ filter on a whole executable; ther
	e's		e's
	no need to apply it only on the executable section. Applyin		no need to apply it only on the executable section. Applying
	g a		a
	BCJ filter on an archive that contains both executable and n		BCJ filter on an archive that contains both executable and n
	on-		on-
	executable files may or may not give good results, so it gen		executable files may or may not give good results, so it gen
	er‐		er‐
	ally isn't good to blindly apply a BCJ filter when compress		ally isn't good to blindly apply a BCJ filter when compress
	ing		ing
	binary packages for distribution.		binary packages for distribution.
	These BCJ filters are very fast and use insignificant amount		These BCJ filters are very fast and use insignificant amount
	of		of
	memory. If a BCJ filter improves compression ratio of a fi		memory. If a BCJ filter improves compression ratio of a fi
	le,		le,
	it can improve decompression speed at the same time. This		it can improve decompression speed at the same time. This
	is		is
	because, on the same hardware, the decompression speed of LZ		because, on the same hardware, the decompression speed of LZ
	MA2		MA2
	is roughly a fixed number of bytes of compressed data per s		is roughly a fixed number of bytes of compressed data per s
	ec‐		ec‐
	ond.		ond.
	These BCJ filters have known problems related to the compress ion		These BCJ filters have known problems related to the compress ion
	ratio:		ratio:
	· Some types of files containing executable code (e.g. obj		· Some types of files containing executable code (e.g. obj
	ect		ect
	files, static libraries, and Linux kernel modules) have		files, static libraries, and Linux kernel modules) have
	the		the
	addresses in the instructions filled with filler valu		addresses in the instructions filled with filler valu
	es.		es.
	These BCJ filters will still do the address conversion, wh ich		These BCJ filters will still do the address conversion, wh ich
	will make the compression worse with these files.		will make the compression worse with these files.
	· Applying a BCJ filter on an archive containing multiple si mi‐		· Applying a BCJ filter on an archive containing multiple si mi‐
	lar executables can make the compression ratio worse than not		lar executables can make the compression ratio worse than not
	using a BCJ filter. This is because the BCJ filter does		using a BCJ filter. This is because the BCJ filter does
	n't		n't
	detect the boundaries of the executable files, and does		detect the boundaries of the executable files, and does
	n't		n't
	reset the address conversion counter for each executable.		reset the address conversion counter for each executable.
	Both of the above problems will be fixed in the future in a		Both of the above problems will be fixed in the future in a
	new		new
	filter. The old BCJ filters will still be useful in embed		filter. The old BCJ filters will still be useful in embed
	ded		ded
	systems, because the decoder of the new filter will be big		systems, because the decoder of the new filter will be big
	ger		ger
	and use more memory.		and use more memory.
	Different instruction sets have have different alignment:		Different instruction sets have have different alignment:
	Filter Alignment Notes		Filter Alignment Notes
	x86 1 32-bit or 64-bit x86		x86 1 32-bit or 64-bit x86
	PowerPC 4 Big endian only		PowerPC 4 Big endian only
	ARM 4 Little endian only		ARM 4 Little endian only
	ARM-Thumb 2 Little endian only		ARM-Thumb 2 Little endian only
	IA-64 16 Big or little endian		IA-64 16 Big or little endian
	SPARC 4 Big or little endian		SPARC 4 Big or little endian
	Since the BCJ-filtered data is usually compressed with LZM		Since the BCJ-filtered data is usually compressed with LZM
	A2,		A2,
	the compression ratio may be improved slightly if the LZ		the compression ratio may be improved slightly if the LZ
	MA2		MA2
	options are set to match the alignment of the selected BCJ f		options are set to match the alignment of the selected BCJ f
	il‐		il‐
	ter. For example, with the IA-64 filter, it's good to set p		ter. For example, with the IA-64 filter, it's good to set p
	b=4		b=4
	with LZMA2 (2^4=16). The x86 filter is an exception; it's u		with LZMA2 (2^4=16). The x86 filter is an exception; it's u
	su‐		su‐
	ally good to stick to LZMA2's default four-byte alignment w		ally good to stick to LZMA2's default four-byte alignment w
	hen		hen
	compressing x86 executables.		compressing x86 executables.
	All BCJ filters support the same options:		All BCJ filters support the same options:
	start=offset		start=offset
	Specify the start offset that is used when convert ing		Specify the start offset that is used when convert ing
	between relative and absolute addresses. The offset m ust		between relative and absolute addresses. The offset m ust
	be a multiple of the alignment of the filter (see the ta‐		be a multiple of the alignment of the filter (see the ta‐
	ble above). The default is zero. In practice,		ble above). The default is zero. In practice,
	the		the
	default is good; specifying a custom offset is alm		default is good; specifying a custom offset is alm
	ost		ost
	never useful.		never useful.
	--delta[=options]		--delta[=options]
	Add the Delta filter to the filter chain. The Delta filter can		Add the Delta filter to the filter chain. The Delta filter can
	be only used as a non-last filter in the filter chain.		be only used as a non-last filter in the filter chain.
	Currently only simple byte-wise delta calculation is support		Currently only simple byte-wise delta calculation is support
	ed.		ed.
	It can be useful when compressing e.g. uncompressed bit		It can be useful when compressing e.g. uncompressed bit
	map		map
	images or uncompressed PCM audio. However, special purp		images or uncompressed PCM audio. However, special purp
	ose		ose
	algorithms may give significantly better results than Delt		algorithms may give significantly better results than Delta
	a +		+
	LZMA2. This is true especially with audio, which compres		LZMA2. This is true especially with audio, which compres
	ses		ses
	faster and better e.g. with flac(1).		faster and better e.g. with flac(1).
	Supported options:		Supported options:
	dist=distance		dist=distance
	Specify the distance of the delta calculation in byt es.		Specify the distance of the delta calculation in byt es.
	distance must be 1-256. The default is 1.		distance must be 1-256. The default is 1.
	For example, with dist=2 and eight-byte input A1 B1 A2 B3		For example, with dist=2 and eight-byte input A1 B1 A2 B3
	A3 B5 A4 B7, the output will be A1 B1 01 02 01 02 01 0 2.		A3 B5 A4 B7, the output will be A1 B1 01 02 01 02 01 0 2.
	Other options		Other options
	-q, --quiet		-q, --quiet
	Suppress warnings and notices. Specify this twice to suppr ess		Suppress warnings and notices. Specify this twice to suppr ess
	errors too. This option has no effect on the exit status. T hat		errors too. This option has no effect on the exit status. T hat
	is, even if a warning was suppressed, the exit status to in di‐		is, even if a warning was suppressed, the exit status to in di‐
	cate a warning is still used.		cate a warning is still used.
	-v, --verbose		-v, --verbose
	Be verbose. If standard error is connected to a terminal,		Be verbose. If standard error is connected to a terminal,
	xz		xz
	will display a progress indicator. Specifying --verbose tw		will display a progress indicator. Specifying --verbose tw
	ice		ice
	will give even more verbose output.		will give even more verbose output.
	The progress indicator shows the following information:		The progress indicator shows the following information:
	· Completion percentage is shown if the size of the input f ile		· Completion percentage is shown if the size of the input f ile
	is known. That is, the percentage cannot be shown in pipe s.		is known. That is, the percentage cannot be shown in pipe s.
	· Amount of compressed data produced (compressing) or consu med		· Amount of compressed data produced (compressing) or consu med
	(decompressing).		(decompressing).
	· Amount of uncompressed data consumed (compressing) or p ro‐		· Amount of uncompressed data consumed (compressing) or p ro‐
	duced (decompressing).		duced (decompressing).
	· Compression ratio, which is calculated by dividing the amo unt		· Compression ratio, which is calculated by dividing the amo unt
	of compressed data processed so far by the amount of unc om‐		of compressed data processed so far by the amount of unc om‐
	pressed data processed so far.		pressed data processed so far.
	· Compression or decompression speed. This is measured as		· Compression or decompression speed. This is measured as
	the		the
	amount of uncompressed data consumed (compression) or p		amount of uncompressed data consumed (compression) or p
	ro‐		ro‐
	duced (decompression) per second. It is shown after a		duced (decompression) per second. It is shown after a
	few		few
	seconds have passed since xz started processing the file.		seconds have passed since xz started processing the file.
	· Elapsed time in the format M:SS or H:MM:SS.		· Elapsed time in the format M:SS or H:MM:SS.
	· Estimated remaining time is shown only when the size of		· Estimated remaining time is shown only when the size of
	the		the
	input file is known and a couple of seconds have alre		input file is known and a couple of seconds have alre
	ady		ady
	passed since xz started processing the file. The time		passed since xz started processing the file. The time
	is		is
	shown in a less precise format which never has any colo		shown in a less precise format which never has any colo
	ns,		ns,
	e.g. 2 min 30 s.		e.g. 2 min 30 s.
	When standard error is not a terminal, --verbose will make xz		When standard error is not a terminal, --verbose will make xz
	print the filename, compressed size, uncompressed size, compr es‐		print the filename, compressed size, uncompressed size, compr es‐
	sion ratio, and possibly also the speed and elapsed time on a		sion ratio, and possibly also the speed and elapsed time o n a
	single line to standard error after compressing or decompress ing		single line to standard error after compressing or decompress ing
	the file. The speed and elapsed time are included only when the		the file. The speed and elapsed time are included only when the
	operation took at least a few seconds. If the operation did		operation took at least a few seconds. If the operation did
	n't		n't
	finish, e.g. due to user interruption, also the completion p		finish, e.g. due to user interruption, also the completion p
	er‐		er‐
	centage is printed if the size of the input file is known.		centage is printed if the size of the input file is known.
	-Q, --no-warn		-Q, --no-warn
	Don't set the exit status to 2 even if a condition worth a wa rn‐		Don't set the exit status to 2 even if a condition worth a wa rn‐
	ing was detected. This option doesn't affect the verbos		ing was detected. This option doesn't affect the verbos
	ity		ity
	level, thus both --quiet and --no-warn have to be used to		level, thus both --quiet and --no-warn have to be used to
	not		not
	display warnings and to not alter the exit status.		display warnings and to not alter the exit status.
	--robot		--robot
	Print messages in a machine-parsable format. This is inten		Print messages in a machine-parsable format. This is inten
	ded		ded
	to ease writing frontends that want to use xz instead		to ease writing frontends that want to use xz instead
	of		of
	liblzma, which may be the case with various scripts. The out put		liblzma, which may be the case with various scripts. The out put
	with this option enabled is meant to be stable across xz		with this option enabled is meant to be stable across xz
	releases. See the section ROBOT MODE for details.		releases. See the section ROBOT MODE for details.
	--info-memory		--info-memory
	Display, in human-readable format, how much physical mem		Display, in human-readable format, how much physical mem
	ory		ory
	(RAM) xz thinks the system has and the memory usage limits		(RAM) xz thinks the system has and the memory usage limits
	for		for
	compression and decompression, and exit successfully.		compression and decompression, and exit successfully.
	-h, --help		-h, --help
	Display a help message describing the most commonly u sed		Display a help message describing the most commonly u sed
	options, and exit successfully.		options, and exit successfully.
	-H, --long-help		-H, --long-help
	Display a help message describing all features of xz, and e xit		Display a help message describing all features of xz, and e xit
	successfully		successfully
	-V, --version		-V, --version
	Display the version number of xz and liblzma in human reada		Display the version number of xz and liblzma in human reada
	ble		ble
	format. To get machine-parsable output, specify --robot bef		format. To get machine-parsable output, specify --robot bef
	ore		ore
	--version.		--version.
	ROBOT MODE		ROBOT MODE
	The robot mode is activated with the --robot option. It makes the o ut‐		The robot mode is activated with the --robot option. It makes the o ut‐
	put of xz easier to parse by other programs. Currently --robot is s up‐		put of xz easier to parse by other programs. Currently --robot is s up‐
	ported only together with --version, --info-memory, and --list.		ported only together with --version, --info-memory, and --list.
	It		It
	will be supported for normal compression and decompression in		will be supported for compression and decompression in the future.
	the
	future.
	Version		Version
	xz --robot --version will print the version number of xz and liblzma in		xz --robot --version will print the version number of xz and liblzma in
	the following format:		the following format:
	XZ_VERSION=XYYYZZZS		XZ_VERSION=XYYYZZZS
	LIBLZMA_VERSION=XYYYZZZS		LIBLZMA_VERSION=XYYYZZZS
	X Major version.		X Major version.
	skipping to change at line 1210		skipping to change at line 1283
	NOTES		NOTES
	Compressed output may vary		Compressed output may vary
	The exact compressed output produced from the same uncompressed in put		The exact compressed output produced from the same uncompressed in put
	file may vary between XZ Utils versions even if compression options are		file may vary between XZ Utils versions even if compression options are
	identical. This is because the encoder can be improved (faster or b et‐		identical. This is because the encoder can be improved (faster or b et‐
	ter compression) without affecting the file format. The output can		ter compression) without affecting the file format. The output can
	vary even between different builds of the same XZ Utils version, if		vary even between different builds of the same XZ Utils version, if
	different build options are used.		different build options are used.
	The above means that implementing --rsyncable to create rsyncable		The above means that once --rsyncable has been implemented, the resu
	.xz		lt‐
	files is not going to happen without freezing a part of the enco		ing files won't necessarily be rsyncable unless both old and new fi
	der		les
	implementation, which can then be used with --rsyncable.		have been compressed with the same xz version. This problem can
			be
			fixed if a part of the encoder implementation is frozen to keep rsyn
			ca‐
			ble output stable across xz versions.
	Embedded .xz decompressors		Embedded .xz decompressors
	Embedded .xz decompressor implementations like XZ Embedded don't nec es‐		Embedded .xz decompressor implementations like XZ Embedded don't nec es‐
	sarily support files created with integrity check types other than n one		sarily support files created with integrity check types other than n one
	and crc32. Since the default is --check=crc64, you must use		and crc32. Since the default is --check=crc64, you must use
	--check=none or --check=crc32 when creating files for embedded syste ms.		--check=none or --check=crc32 when creating files for embedded syste ms.
	Outside embedded systems, all .xz format decompressors support all the		Outside embedded systems, all .xz format decompressors support all the
	check types, or at least are able to decompress the file without ve ri‐		check types, or at least are able to decompress the file without ve ri‐
	fying the integrity check if the particular check is not supported.		fying the integrity check if the particular check is not supported.
	skipping to change at line 1310		skipping to change at line 1385
	-0 ... -9 and --extreme are useful when customizing LZMA2 prese ts.		-0 ... -9 and --extreme are useful when customizing LZMA2 prese ts.
	Here are the relevant parts collected from those two tables:		Here are the relevant parts collected from those two tables:
	Preset CompCPU		Preset CompCPU
	-0 0		-0 0
	-1 1		-1 1
	-2 2		-2 2
	-3 3		-3 3
	-4 4		-4 4
	-5 5		-5 5
	-6 6		-6 6
	-5e 7		-5e 7
	-6e 8		-6e 8
	If you know that a file requires somewhat big dictionary (e.g. 32 M iB)		If you know that a file requires somewhat big dictionary (e.g. 32 M iB)
	to compress well, but you want to compress it quicker than xz -8 wo uld		to compress well, but you want to compress it quicker than xz -8 wo uld
	do, a preset with a low CompCPU value (e.g. 1) can be modified to us e a		do, a preset with a low CompCPU value (e.g. 1) can be modified to us e a
	bigger dictionary:		bigger dictionary:
	xz --lzma2=preset=1,dict=32MiB foo.tar		xz --lzma2=preset=1,dict=32MiB foo.tar
	With certain files, the above command may be faster than xz -6 wh ile		With certain files, the above command may be faster than xz -6 wh ile
	skipping to change at line 1398		skipping to change at line 1473
	number of bytes per pixel.		number of bytes per pixel.
	SEE ALSO		SEE ALSO
	xzdec(1), xzdiff(1), xzgrep(1), xzless(1), xzmore(1), gzip( 1),		xzdec(1), xzdiff(1), xzgrep(1), xzless(1), xzmore(1), gzip( 1),
	bzip2(1), 7z(1)		bzip2(1), 7z(1)
	XZ Utils: <http://tukaani.org/xz/>		XZ Utils: <http://tukaani.org/xz/>
	XZ Embedded: <http://tukaani.org/xz/embedded.html>		XZ Embedded: <http://tukaani.org/xz/embedded.html>
	LZMA SDK: <http://7-zip.org/sdk.html>		LZMA SDK: <http://7-zip.org/sdk.html>
	Tukaani 2012-07-03 XZ (1)		Tukaani 2014-12-16 XZ (1)
End of changes. 85 change blocks.
	316 lines changed or deleted		439 lines changed or added
This html diff was produced by rfcdiff 1.41. The latest version is available from http://tools.ietf.org/tools/rfcdiff/